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: [RFA] remove more df_byte_scan corpses

From: Paolo Bonzini <bonzini at gnu dot org>
To: Bernd Schmidt <bernds at codesourcery dot com>
Cc: GCC Patches <gcc-patches at gcc dot gnu dot org>
Date: Tue, 17 Aug 2010 02:13:23 +0200
Subject: Re: [RFA] remove more df_byte_scan corpses
References: <4C69BC3B.9040001@gnu.org> <4C69D1DF.5030403@codesourcery.com>

On 08/17/2010 02:03 AM, Bernd Schmidt wrote:

On 08/17/2010 12:31 AM, Paolo Bonzini wrote:
The whole DF_REF_EXTRACT machinery can be eliminated, and only the flags
left in for the benefit of IRA.
Bootstrapped/regtested x86_64-pc-linux-gnu, ok?
Looks ok to me.

I wonder if there's any point to DF_REF_*_EXTRACT

IRA uses it, I don't know it well enough to understand why. I thought about simplifying DF_REF_SIGN_EXTRACT vs. DF_REF_ZERO_EXTRACT, but since there is also DF_REF_STRICT_LOW_PART (and IRA uses it too, distinguishing DF_REF_ZERO_EXTRACT and DF_REF_STRICT_LOW_PART) it would be incoherent and generally it would probably lose readability.

and why there's a need to have both DF_REF_PARTIAL and DF_REF_READ_WRITE?

DF_REF_READ_WRITE is only set for subregs which appear on the LHS of a set.

Paolo

References:
- [RFA] remove more df_byte_scan corpses
  - From: Paolo Bonzini
- Re: [RFA] remove more df_byte_scan corpses
  - From: Bernd Schmidt

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