This is the mail archive of the
mailing list for the GDB project.
Re: [RFA 2/4] dwarf2_physname
- From: André Pönitz <andre dot poenitz at nokia dot com>
- To: gdb-patches at sourceware dot org
- Date: Mon, 23 Nov 2009 08:26:59 +0100
- Subject: Re: [RFA 2/4] dwarf2_physname
- References: <4B0707E7.firstname.lastname@example.org>
On Friday 20 November 2009 22:19:35 Keith Seitz wrote:
> [...] Some warnings are in order. First, this is probably going to really slow
> down large C++ applications, because we no longer use
> DW_AT_MIPS_linkage_name as generated from the compiler. We essentially
> build this name during DIE reading. [...]
> With those caveats out of the way, I will say that I understand there
> are some controversial changes in here: it is a pretty big paradigm
> shift for the dwarf reader. Nonetheless, the benefits outweigh the
> performance degradation (for C++ ONLY) and the minor java regression.
> IMO, the elimination of single-quoting for C++ expressions alone is
> worth it.
I am really worried about the performance degradation you expect.
How bad will that be?
Medium to large sized C++ applications is basically "all I have", and speed
is already by now the biggest issue I have with gdb. 
If the deal is "correct but slow" vs "flaky but faster" I surely prefer the
"flaky but fast" side. I have encountered quite a few issues with C++ in
gdb, so yes, that's not nice. But usually one can work around somehow on
the user side. Raw speed on the other hand is nothing the user can improve.
 Actually "the biggest issue that is left": Python scripting turned out
to be a panacea for basically everything else. Thanks to everybody involved.