Re: [PATCH] scripts/gdb: fix lx-version for gdb 7.3-

From: Kieran Bingham
Date: Wed Oct 17 2018 - 08:31:34 EST

Hi Du,

On 17/10/18 03:36, Du Changbin wrote:
> For gdb version less than 7.3, lx-version only one character.
> (gdb) lx-version
> L(gdb)
> This can be fixed by casting 'linux_banner' as (char *).
> (gdb) lx-version
> Linux version 4.19.0-rc1+ (changbin@acer) (gcc version 7.3.0 (Ubuntu 7.3.0-16ubuntu3)) #21 SMP Sat Sep 1 21:43:30 CST 2018
> gdb 7.4 seems to be no such issue.
> Signed-off-by: Du Changbin <changbin.du@xxxxxxxxx>
> ---
> scripts/gdb/linux/ | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
> diff --git a/scripts/gdb/linux/ b/scripts/gdb/linux/
> index 086d27223c0c..0aebd7565b03 100644
> --- a/scripts/gdb/linux/
> +++ b/scripts/gdb/linux/
> @@ -41,7 +41,7 @@ class LxVersion(gdb.Command):
> def invoke(self, arg, from_tty):
> # linux_banner should contain a newline
> - gdb.write(gdb.parse_and_eval("linux_banner").string())
> + gdb.write(gdb.parse_and_eval("(char *)linux_banner").string())

Interesting - this is an issue I reported to the GDB team after I'd
tried to dig into the fault.

the parse_and_eval finds a string length of 0, but then later in the GBD
parsing it gets incremented to 1, hence the single char. I guess it's
some failure in GDB determining the type of the pointer.

However, from reading the bugzilla entry and from what I recall - it
only occurred if you had done a backtrace before calling lx-version.

Did you discover that it would always return only the single character
in your testing ? Even if it was the first command executed?

My worry is - that if this fixes lx-version - do we also need to
manually cast lx-cmdline as well ? (I never saw the issue occur on that

I think considering the age of the bugzilla at GDB, and the lack of any
GDB level fix or understanding - we should certainly apply this fix here
- and probably include it in the stable trees too.

With the title fix as highlighted by Jan,

Reviewed-by: Kieran Bingham <kbingham@xxxxxxxxxx>

Would you be able to test the issue with the lx-cmdline command as well
please? And if it is suffering the same issue - perhaps you could either
submit a second patch to update that entry - or respin this one to
include both?

> LxVersion()