GDB 10 branching - 2020-06-21 Update

[Andrew Burgess]

* Joel Brobecker <brobecker@adacore.com> [2020-06-21 10:57:29 -0700]:

> Hi everyone,
> 
> Below is the current status of all items we want to look at before
> branching. Please let me know if there are other issues you think
> are worth adding to the list.
> 
> Thank you!
> 
> Fixed Since the Previous Update:
> --------------------------------
> 
>   * [PedroA]
>     <PR gdb/25412> thread_info with duplicate ptid added to inferior thread list
>     https://sourceware.org/bugzilla/show_bug.cgi?id=25412
> 
>   * Check exec-file has changed before checking for exec-file mismatch
>     https://sourceware.org/pipermail/gdb-patches/2020-June/169287.html
> 
> Added Since the Last Update:
> ----------------------------
> 
>   * [TomT/HannesD]
>     <PR win32/25302> Mismatching fstat() function calls in gdb_bfd_open() and cache_bstat()
>     https://sourceware.org/bugzilla/show_bug.cgi?id=25302
> 
>     Latest discussion at:
>     https://sourceware.org/pipermail/gdb-patches/2020-June/169670.html
> 
> Other Ongoing Items:
> --------------------
> 
>   * [RainerO]
>     <PR gdb/25939> [10 regression] run fails with ICE on Solaris
>     https://sourceware.org/bugzilla/show_bug.cgi?id=25939
> 
>     Being discussed at:
>     https://sourceware.org/pipermail/gdb-patches/2020-June/169554.html
> 
>     (one?) Issue identified by Pedro, and patch proposed that fixes
>     the identified issue -- unpushed as of yet, I believe:
>     https://sourceware.org/pipermail/gdb-patches/2020-June/169674.html
> 
>     Is that sufficient to close the PR?
>     Can we consider the Solaris port functional again?
> 
>   * [AndrewB/TomT]
>     QEMU / GDB compatibility on RISCV64 ELF (failure to fetch some registers)
>     https://sourceware.org/pipermail/gdb-patches/2020-June/169393.html
> 
>     New patch series (expected to be pushed this coming week):
>     https://sourceware.org/pipermail/gdb-patches/2020-June/169561.html
>     RISC-V target description and register handling fixes
> 
>   * [PhilippeW]
>     <PR gdb/25475>
>     https://sourceware.org/bugzilla/show_bug.cgi?id=25475
> 
>     Seems like a minor bug, but current thinking was that resolution
>     requires a change of behavior. I made some proposals, one of them
>     could hopefully be done without triggering a change of behavior.
>     But, IMO, this discussion deserves a wider audience, so I suggested
>     a thread be started on gdb-patches. Philippe answered he'll propose
>     a patch.
> 
> Not Critical, but Requested:
> ----------------------------
> 
>   * [PhilippeW]
>   Allow the user to define default args for aliases

I'd like to suggest that we also wait for a fix to the issue discussed
here:

  https://sourceware.org/pipermail/gdb-patches/2020-June/169679.html

with a demo for the actual bug in this mail:

  https://sourceware.org/pipermail/gdb-patches/2020-June/169681.html

This is a regression for a corner case of the tail-call unwinder and
inline-frame unwinder in combination.  I have a fix for this issue
which I'm testing locally and will post later today.

It's going to appear as a replacement for this patch:

  https://sourceware.org/pipermail/gdb-patches/2020-June/169621.html

Which is #5 in a series, however, the series has already been
reviewed, and other than a few small nits is ready to merge.  The new
patch #5 will need looking over though, and might cause some
discussion.

As Tom has pointed out, the fix in patch #5 will also close PR
python/22748, breakage in Python frame unwinders, which is a nice win
too.

Thanks,
Andrew