Re: [PATCH 0/4] Enable the user to dump all memory regions

[Sergio Durigan Junior <sergiodj at redhat dot com>]

On Tuesday, November 28 2017, Sergio Lopez wrote:

> GDB versions prior to df8411da087dc05481926f4c4a82deabc5bc3859
> unconditionally included all memory regions in the core dump.
>
> After that change, while is still possible to ask GDB to ignore
> /proc/PID/coredump_filter using the 'set use-coredump-filter' command,
> there's no way to request it to dump regions marked with the VM_DONTDUMP
> flag ("dd" in /proc/PID/smaps").
>
> This patch series implement the new 'set honor-dontdump-flag' command
> for GDB, and the "-a" argument for gcore, allowing the user to mimic the
> behavior of previous GDB versions.

Thanks for the patch, Sergio.

Overall I agree with the approach, but I've made a few comments here and
there about things I think should be addressed.  The most important of
them is the command name.

I would really like to see a test for this.  We already carry a testcase
for the use-coredump-filter command (see gdb.base/coredump-filter.exp),
so I think you can just extend it to test this specific feature.

Thanks,

> Sergio Lopez (4):
>   Implement 'set honor-dontdump-flag' command
>   Document new 'set honor-dontdump-flag' command
>   Add "-a" argument to gcore.in
>   Document the new "-a" argument for gcore
>
>  gdb/ChangeLog       | 10 ++++++++++
>  gdb/doc/ChangeLog   |  9 +++++++++
>  gdb/doc/gdb.texinfo | 18 +++++++++++++++++-
>  gdb/gcore.in        | 25 +++++++++++++++++++++++--
>  gdb/linux-tdep.c    | 19 ++++++++++++++++++-
>  5 files changed, 77 insertions(+), 4 deletions(-)
>
> -- 
> 2.14.3

-- 
Sergio
GPG key ID: 237A 54B1 0287 28BF 00EF  31F4 D0EB 7628 65FC 5E36
Please send encrypted e-mail if possible
http://sergiodj.net/