* [FYI]: Incoming C++ changes
@ 2000-06-05 13:52 Daniel Berlin
0 siblings, 0 replies; 2+ messages in thread
From: Daniel Berlin @ 2000-06-05 13:52 UTC (permalink / raw)
To: gdb-patches; +Cc: gdb
I made quite a bunch of C++ improvements over the weekend (i was gone
until 3pm today), i'll be committing in a moment.
In no particular order:
1. Symbol table performance is vastly improved. No longer do we need to
force a linear search for C++, ever.
This required simply changing STREQ(SYMBOL_NAME,name)) to
SYMBOL_MATCHES_NAME. Once that was done, the issues that required doing
the linear search in the first place no longer occurred. This change
causes no regressions.
2. Template handling improvements. Should be as good as it is on HP now.
Required removing some code in the expression parser that was dupicating,
but doing less, than something that could already do it.
3. Test suite improvements: Tests we previously skipped on g++, we now
don't skip, and pass (particularly, template tests). I also added a
namespace test from the gdb.hp dir. This requirred changing a few tests to
accept valid output.
This actually exposes what appears to be a bug in demangling when you use
stabs.
^ permalink raw reply [flat|nested] 2+ messages in thread
* [FYI]: Incoming C++ changes
@ 2000-06-05 13:51 Daniel Berlin
0 siblings, 0 replies; 2+ messages in thread
From: Daniel Berlin @ 2000-06-05 13:51 UTC (permalink / raw)
To: gdb-patches; +Cc: gdb
I made quite a bunch of C++ improvements over the weekend (i was gone
until 3pm today), i'll be committing in a moment.
In no particular order:
1. Symbol table performance is vastly improved. No longer do we need to
force a linear search for C++, ever.
This required simply changing STREQ(SYMBOL_NAME,name)) to
SYMBOL_MATCHES_NAME. Once that was done, the issues that required doing
the linear search in the first place no longer occurred. This change
causes no regressions.
2. Template handling improvements. Should be as good as it is on HP now.
Required removing some code in the expression parser that was dupicating,
but doing less, than something that could already do it.
3. Test suite improvements: Tests we previously skipped on g++, we now
don't skip, and pass (particularly, template tests). I also added a
namespace test from the gdb.hp dir. This requirred changing a few tests to
accept valid output.
This actually exposes what appears to be a bug in demangling when you use
stabs.
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2000-06-05 13:52 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2000-06-05 13:52 [FYI]: Incoming C++ changes Daniel Berlin
-- strict thread matches above, loose matches on Subject: below --
2000-06-05 13:51 Daniel Berlin
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).