Mirror of the gdb-patches mailing list
 help / color / mirror / Atom feed
From: Daniel Berlin <dan@cgsoftware.com>
To: Jim Blandy <jimb@zwingli.cygnus.com>
Cc: Charlie Mills <cmills@synopsys.com>, gdb-patches@sourceware.cygnus.com
Subject: Re: Simple but crucial bug fix to gdb
Date: Wed, 30 May 2001 21:23:00 -0000	[thread overview]
Message-ID: <871yp6tael.fsf@dynamic-addr-83-177.resnet.rochester.edu> (raw)
In-Reply-To: <npwv6ymcrj.fsf@zwingli.cygnus.com>

Jim Blandy <jimb@zwingli.cygnus.com> writes:

> > I would like to submit a very simple patch to gdb.
> > I would like if possible to avoid legal issues (avoid having to
> > submit a form to our legal department) by simply describing the fix,
> > which is a diff of only a few characters, rather than sending you a
> > source file.
> 
> The size of the text you send doesn't make any difference --- it's the
> size of the change that matters.  You could have posted an ordinary
> patch without legal troubles.
> 
> I find it odd that the stabs reader is seeing a function definition
> symbol (a descriptor of 'f') before it's seen anything to set up the
> psymtab.  Does your object file really have an N_FUN stab (or
> whatever) before the first N_SO stab?

Ha.
You are so funny.
Your joking, right?
This is the same STABS reader keeps psymtabs around so that we can't free them
after conversion to symtabs, at the expense of every other debug info
reader. And requires hacks in buildsym that only the other 8 year old
symbol reader (mdebugread, split off from mipsread) requires?
You are surprised it does something else odd?
Anyway, ranting aside, you are completely correct in this case.
The first symbol is *either* the GCC_COMPILED_FLAG_SYMBOL (followed
directly by an N_SO), or the N_SO.
It says so in read_ofile_symtab, in very explicit terms ("The N_SO
starting this symtab is the first symbol, so we better not check the
symbol before it")

So either all the comments are wrong, or gcc is broken.
> 
> If we just ignore the N_FUN, I'm worried that the pst's textlow will
> not be set correctly --- that is, I'm not sure it's safe to just
> ignore the N_FUN when pst is zero.
> 
> Could you post the result of running `objdump --stabs' on your
> executable, or object file, if it's not too large or revealing?

-- 
"We were in Salino, Utah when we were arrested for not going
through a green light.  We pleaded "maybe".  I asked the judge
if he knew what time it is, he did, and I said, "No further
questions."
"-Steven Wright


  reply	other threads:[~2001-05-30 21:23 UTC|newest]

Thread overview: 26+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2001-05-30 14:27 Charlie Mills
2001-05-30 14:36 ` Christopher Faylor
2001-05-30 15:16   ` Fernando Nasser
2001-05-30 20:18     ` Jim Blandy
2001-05-30 20:25       ` Christopher Faylor
2001-05-31  5:30       ` Fernando Nasser
2001-05-31 13:55         ` Jim Blandy
2001-05-31 14:43           ` Fernando Nasser
2001-05-31 16:46           ` Christopher Faylor
2001-05-31 17:40             ` Daniel Berlin
2001-05-31 20:00               ` Frank Ch. Eigler
2001-06-01  9:41                 ` Fernando Nasser
2001-06-01 10:01                 ` Michael Snyder
2001-06-01 11:14                   ` Daniel Berlin
2001-06-01 11:25                     ` Fernando Nasser
2001-06-01 14:05                       ` Charlie Mills
2001-06-06  6:01                 ` Andrew Cagney
2001-05-30 20:00   ` Christopher Faylor
2001-05-30 20:43     ` Christopher Faylor
2001-05-30 20:12 ` Jim Blandy
2001-05-30 21:23   ` Daniel Berlin [this message]
2001-06-01 13:35 ` Jim Blandy
2001-06-01 13:41   ` Fernando Nasser
2001-06-01 14:07   ` Daniel Berlin
2001-06-01 14:15     ` Jim Blandy
2001-06-01 14:17     ` Jim Blandy

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=871yp6tael.fsf@dynamic-addr-83-177.resnet.rochester.edu \
    --to=dan@cgsoftware.com \
    --cc=cmills@synopsys.com \
    --cc=gdb-patches@sourceware.cygnus.com \
    --cc=jimb@zwingli.cygnus.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox