This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]
Other format:	[Raw text]

Re: [PATCH][gcse] Use emit_move_insn rather than creating SET rtx and emitting that

From: Kyrill Tkachov <kyrylo dot tkachov at arm dot com>
To: Steven Bosscher <stevenb dot gcc at gmail dot com>
Cc: GCC Patches <gcc-patches at gcc dot gnu dot org>, Jakub Jelinek <jakub at redhat dot com>, James Greenhalgh <James dot Greenhalgh at arm dot com>
Date: Thu, 19 Mar 2015 14:58:16 +0000
Subject: Re: [PATCH][gcse] Use emit_move_insn rather than creating SET rtx and emitting that
Authentication-results: sourceware.org; auth=none
References: <550AD2EF dot 30709 at arm dot com> <CABu31nPze9xcEu8tbKtuchxGgxKu0VhhxY8RD+sF8YB0U8MAJw at mail dot gmail dot com>


On 19/03/15 13:56, Steven Bosscher wrote:

On Thu, Mar 19, 2015 at 2:45 PM, Kyrill Tkachov wrote:

As pointed out by James Greenhalgh offline the correct thing would have been
to do an
emit_move_insn to let the backend expanders do the right thing (especially
in the concerned
testcase gcc.c-torture/execute/pr65427.c that uses 256-bit vectors that arm
doesn't support
natively).

This is supposed to be caught by want_to_gcse_p() via
can_assign_to_reg_without_clobbers_p(). How does your expression get
past that barrier?

The gcc_unreachable() is there because all the code in gcse.c assumes
it is OK to emit a SET-insn without going through emit_move_insn().


btw, the ICE happens during the hoist pass, rather than gcse.
Kyrill

Ciao!
Steven

References:
- [PATCH][gcse] Use emit_move_insn rather than creating SET rtx and emitting that
  - From: Kyrill Tkachov
- Re: [PATCH][gcse] Use emit_move_insn rather than creating SET rtx and emitting that
  - From: Steven Bosscher

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]