Mirror of the gdb-patches mailing list
 help / color / mirror / Atom feed
From: Doug Evans <dje@google.com>
To: Jan Kratochvil <jan.kratochvil@redhat.com>
Cc: gdb-patches@sourceware.org, tromey@redhat.com, palves@redhat.com
Subject: Re: [RFA] Fix inconsistency in blockvector addrmap vs non-addrmap handling
Date: Fri, 22 Jun 2012 20:51:00 -0000	[thread overview]
Message-ID: <CADPb22QB0czUev9oKXv9LHHK6H1T=8JhFa0whErojZC+cdrdBw@mail.gmail.com> (raw)
In-Reply-To: <20120622193539.GA7344@host2.jankratochvil.net>

On Fri, Jun 22, 2012 at 12:35 PM, Jan Kratochvil
<jan.kratochvil@redhat.com> wrote:
> On Tue, 05 Jun 2012 03:14:46 +0200, Doug Evans wrote:
>> --- buildsym.c        29 May 2012 20:23:17 -0000      1.97
>> +++ buildsym.c        5 Jun 2012 00:26:01 -0000
>> @@ -1024,8 +1027,15 @@ end_symtab (CORE_ADDR end_addr, struct o
>>      {
>>        /* Define the STATIC_BLOCK & GLOBAL_BLOCK, and build the
>>           blockvector.  */
>> -      finish_block (0, &file_symbols, 0, last_source_start_addr,
>> -                 end_addr, objfile);
>> +      struct block *static_block;
>> +
>> +      static_block = finish_block (0, &file_symbols, 0,
>> +                                last_source_start_addr, end_addr,
>> +                                objfile);
>> +      /* Mark the range of the static block so that if we end up using
>> +      blockvector.map then find_block_in_blockvector behaves identically
>> +      regardless of whether the addrmap is present.  */
>> +      record_block_range (static_block, last_source_start_addr, end_addr - 1);
>
> On IRC Doug made a note:
>        Arguably the second is the better fix but it's still a hack as
>        addrmaps are intended to handle discontiguous symtabs and this defeats
>        that.
>
> Where "the first fix" was:
>        [RFA] Fix gdb segv in dw2_find_pc_sect_symtab
>        http://sourceware.org/ml/gdb-patches/2012-05/msg00958.html
>
> I think the right way is to call dwarf2_record_block_ranges for
> DW_AT_compilation_unit but I haven't tried to write such patch yet, is there
> a problem?
>
>
>>        finish_block_internal (0, &global_symbols, 0, last_source_start_addr,
>>                            end_addr, objfile, 1);
>>        blockvector = make_blockvector (objfile);

With some work that could be made to work.
But remember that the static block  (or global block) doesn't exist
until the call to end_symtab, and, barring even more work, you need to
get the range into pending_addrmap before the call to make_blockvector
where pending_addrmap is turned into a fixed addrmap.


  reply	other threads:[~2012-06-22 20:51 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-06-05  1:15 Doug Evans
2012-06-05 15:06 ` Pedro Alves
2012-06-06 15:54 ` Jan Kratochvil
2012-06-22 18:24 ` ping: " Jan Kratochvil
2012-06-22 19:36 ` Jan Kratochvil
2012-06-22 20:51   ` Doug Evans [this message]
2012-06-24 18:34     ` Jan Kratochvil
2012-06-25 11:39       ` Doug Evans
2012-07-03 12:54         ` Jan Kratochvil
2012-07-12  5:19           ` Doug Evans
2012-07-12 20:30             ` Jan Kratochvil
2012-07-13 20:28               ` [commit] " Jan Kratochvil

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='CADPb22QB0czUev9oKXv9LHHK6H1T=8JhFa0whErojZC+cdrdBw@mail.gmail.com' \
    --to=dje@google.com \
    --cc=gdb-patches@sourceware.org \
    --cc=jan.kratochvil@redhat.com \
    --cc=palves@redhat.com \
    --cc=tromey@redhat.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