public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
From: "meadori at codesourcery dot com" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug target/58595] internal compiler error: in gen_movsi when compiling on arm some files of lttng-tools with -fPIE
Date: Fri, 21 Feb 2014 19:16:00 -0000	[thread overview]
Message-ID: <bug-58595-4-3AheGyVDUF@http.gcc.gnu.org/bugzilla/> (raw)
In-Reply-To: <bug-58595-4@http.gcc.gnu.org/bugzilla/>

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58595

--- Comment #3 from Meador Inge <meadori at codesourcery dot com> ---
As shown in the backtrace, the ICE occurs when GCSE tries generate a move.
The the following asserts fire in 'gen_movsi':

  gcc_assert (GET_CODE (tmp) == SYMBOL_REF);                                
  gcc_assert (SYMBOL_REF_TLS_MODEL (tmp) != 0);

This is because the move operands look like:

  (gdb) p debug_rtx(operands[0])
  (reg/f:SI 115)

  (gdb) p debug_rtx(operands[1])
  (mem/u/c:SI (plus:SI (reg:SI 111)
          (unspec:SI [
                  (symbol_ref:SI ("state") [flags 0x90] <var_decl 0xf7c50114
state>)
              ] UNSPEC_PIC_SYM)) [0  S4 A32])

We aren't expecting the 'mem' expression with the TLS reference in side.

I am not sure if this is just an overzealous assert or we have a deeper
issue.  Removing the asserts does allow the code to compile.


  parent reply	other threads:[~2014-02-21 19:16 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-10-02 19:12 [Bug target/58595] New: " yannick.brosseau at gmail dot com
2013-10-31  9:49 ` [Bug target/58595] " ramana at gcc dot gnu.org
2013-10-31 10:00 ` ramana at gcc dot gnu.org
2014-02-21 19:13 ` meadori at codesourcery dot com
2014-02-21 19:16 ` meadori at codesourcery dot com [this message]
2014-03-03 13:40 ` jakub at gcc dot gnu.org
2014-03-03 15:47 ` meadori at codesourcery dot com
2014-03-03 16:00 ` jakub at gcc dot gnu.org
2014-03-03 16:12 ` jakub at gcc dot gnu.org
2014-03-03 16:54 ` jakub at gcc dot gnu.org
2014-03-03 17:16 ` ktkachov at gcc dot gnu.org
2014-03-04 10:15 ` ktkachov at gcc dot gnu.org
2014-03-04 13:13 ` jakub at gcc dot gnu.org
2014-03-05  8:59 ` ktkachov at gcc dot gnu.org
2014-03-06 12:07 ` jakub at gcc dot gnu.org
2014-04-10  7:46 ` jakub at gcc dot gnu.org
2014-04-10  8:03 ` jakub at gcc dot gnu.org

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=bug-58595-4-3AheGyVDUF@http.gcc.gnu.org/bugzilla/ \
    --to=gcc-bugzilla@gcc.gnu.org \
    --cc=gcc-bugs@gcc.gnu.org \
    /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;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).