On Mon Jul 6, 2026 at 7:16 AM CEST, Deepesh Varatharajan wrote:
>
> On 05-07-2026 22:59, Yoann Congal wrote:
>> CAUTION: This email comes from a non Wind River email account!
>> Do not click links or open attachments unless you recognize the sender and 
>> know the content is safe.
>>
>> On Tue Jun 30, 2026 at 9:50 AM CEST, Deepesh via lists.openembedded.org 
>> Varatharajan wrote:
>>> From: Deepesh Varatharajan <[email protected]>
>>>
>>> This is a corrective release over GDB 17.1, fixing the following issues:
>>>
>>>   * PR dap/33228 ([gdb/dap] error while listing register children)
>>>
>>>   * PR gdb/33737 (gdb --help says 'For more information, type "stream"
>>>     from within GDB', but "stream" is not a defined command)
>>>
>>>   * PR build/33747 (Incompatible with MUSL libc: no member named 'c_ospeed'
>>>     in 'termios')
>>>
>>>   * PR gdb/33748 (gdb17 regression with displaying ANSI colors)
>>>
>>>   * PR gdb/33753 (Out-of-bounds writes in string_{v}printf -- threads
>>>     and static data don't mix)
>>>
>>>   * PR cli/33761 (Setting style colors is broken on MS-Windows)
>>>
>>>   * PR gdb/33768 (Loading compressed GDB scripts from .debug_gdb_scripts 
>>> fails)
>>>
>>>   * PR symtab/33775 ([gdb/symtab] data race in
>>>     dwarf2_per_cu::{set_addr_size,set_offset_size,set_ref_addr_size})
>>>
>>>   * PR symtab/33777 ([gdb/symtab] dw2_get_file_names doesn't cache result
>>>     for dummy CU)
>>>
>>>   * PR symtab/33825 ([dwz] Extremely slow symbol lookup with DWZ-compressed
>>>     debug info (thousands of partial units))
>>>
>>>   * PR testsuite/33845 (gdb: There are 4 unexpected failures in
>>>     breakpoint-in-ro-region.exp)
>>>
>>>   * PR gdb/33872 (`skip -gfile` has inverted logic)
>>>
>>>   * PR gdb/33926 (GDB 17.1 AArch64: redefinition of user_gcs struct on musl)
>>>
>>>   * PR breakpoints/34112 (rbreak `file:regex` sets breakpoints for matches
>>>     outside of `file` [reproducer attached])
>>>
>>> Drop patches merged upstream:
>>>    * 0009-PR-gdb-33747-gdb-ser-unix-modernize-Linux-custom-bau.patch
>>>    * 0010-GDB-aarch64-linux-Fix-build-failure-on-musl-systems.patch
>>>
>>> Signed-off-by: Deepesh Varatharajan <[email protected]>
>> When looking at the commit log:
>>     $ git log --format=oneline --abbrev-commit 
>> gdb-17.1-release..gdb-17.2-release |grep -v "Automatic date update in 
>> version.in"
>>    7adf9fa6b1c Set GDB version number to 17.2.
>>    063b0d3f0b3 [gdb/breakpoints] Don't ignore <file> in rbreak 
>> <file>:<regexp>
>>    17d05df7da1 gdb/ser-unix: add POSIX cfsetispeed/cfsetospeed support for 
>> custom baud rates
>>    56c4ba5d7b5 gdb/ser-unix: fix musl build failure when setting custom baud 
>> rates
>>    1ccc3f6a2e2 GDB: aarch64-linux: Fix build failure on musl systems
>>    a4477520f93 gdb: Fix unexpected failures in breakpoint-in-ro-region.exp
>>    eaf673ca75c gdb: Make test result accurate in breakpoint-in-ro-region.exp
>>    91080fe0d50 Revert "skip -gfile: call fnmatch without FNM_FILE_NAME"
>>    36356c77b89 [gdb/symtab] Fix slow symbol lookup with dwz
>>    07e542a3bd0 gdb/
>>    101b318a6a6 gdb: fix handling of raw ANSI escape sequences printed from 
>> Python
>>    0ac91006f95 [gdb/symtab] Cache dw2_get_file_names result for dummy CU
>>    32adc7212e4 gdb/dwarf: make dwarf2_per_cu size fields atomic
>>    59ffbefc0ee [gdbsupport] Fix get_print_cell use in threads
>>    0926b8b8339 gdb: fix loading compressed scripts from 
>> `.debug_gdb_scripts`-section
>>    8e246f43277 Change handling of over-long DAP variables requests
>>    4393da33773 Fix --help output
>>    35880717849 Bump GDB's version number to 17.1.90.DATE-git.
>>
>> ... one commit stands out: "17d05df7da1 gdb/ser-unix: add POSIX 
>> cfsetispeed/cfsetospeed support for custom baud rates".
>> This look like feature addition (e.g new config option). But, in this
>> case, if I understood correctly, this is the merged version of the
>> removed patch 
>> "0009-PR-gdb-33747-gdb-ser-unix-modernize-Linux-custom-bau.patch",
>> so this should be okay to merge but I think the commit message must
>> include this explanation.
> Yes, you are right that feature/bug fix is already part of wrynose as a 
> local patch.
> Will explain in commit message and send V3.
>>
>> I wonder why this particular change was not part of the changelog you
>> added...
> It's there in the commit message.
>
> * PR build/33747 (Incompatible with MUSL libc: no member named 'c_ospeed'
>     in 'termios')
>
> The above PR brings in the below commit.
>
> 17d05df7da1 gdb/ser-unix: add POSIX cfsetispeed/cfsetospeed support for 
> custom baud rates

Oh right, the link is no easy to see.
Thanks for the v3, it does look good.

>
> Regards,
> Deepesh

-- 
Yoann Congal
Smile ECS

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#240232): 
https://lists.openembedded.org/g/openembedded-core/message/240232
Mute This Topic: https://lists.openembedded.org/mt/120043565/21656
Group Owner: [email protected]
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[[email protected]]
-=-=-=-=-=-=-=-=-=-=-=-

Reply via email to