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: [PE-POST] Adjust Bit-region in expand_assignment

From: Jeff Law <law at redhat dot com>
To: Bernd Edlinger <bernd dot edlinger at hotmail dot de>, Richard Biener <richard dot guenther at gmail dot com>
Cc: "gcc-patches at gcc dot gnu dot org" <gcc-patches at gcc dot gnu dot org>, Jakub Jelinek <jakub at redhat dot com>
Date: Thu, 05 Dec 2013 21:12:54 -0700
Subject: Re: [PE-POST] Adjust Bit-region in expand_assignment
Authentication-results: sourceware.org; auth=none
References: <DUB122-W396D54115D70C9FF68A8D2E43C0 at phx dot gbl>,<CAFiYyc13uxba37WySOxXf14UVJW15_DM7jggNGbR4H2cK96Eiw at mail dot gmail dot com>,<DUB122-W323D4AAAEA1AFA890BF243E43C0 at phx dot gbl>,<CAFiYyc13wC0jmZ4xELAiL5L5Sbh0yauZWViP2hTdR+H4s0iPSw at mail dot gmail dot com>,<DUB122-W4D571A91831593BA22438E43C0 at phx dot gbl>,<CAFiYyc1ECAdBJTkcziSL_f9SQyYUPY9k19QWqkN+RfLO=i2zaQ at mail dot gmail dot com>,<20130910193228 dot GE6732 at virgil dot suse>,<DUB122-W22DAEEE94B5ED3308F6967E4390 at phx dot gbl>,<CAFiYyc3AHggjJbEmZvC_fh67o+LVpTBRdwyUv8T6EhQFE2vNWQ at mail dot gmail dot com>,<DUB122-W281E996A4885309956418EE4250 at phx dot gbl>,<20130916230945 dot GK6732 at virgil dot suse>,<DUB122-W209385567C3551814F6331E4020 at phx dot gbl>,<DUB122-W44FF2B06EE812E903A9365E4F00 at phx dot gbl> <DUB122-W46DEE578F1C4A1E159AA2FE4D50 at phx dot gbl>,<529EC3D4 dot 5010109 at redhat dot com> <DUB122-W46F8B3DF5A12D1D619017AE4D40 at phx dot gbl>

On 12/04/13 00:49, Bernd Edlinger wrote:


Regarding the consistency of bitregion_start/end ,

they should either both be zero, or

bitregion_start <= bitpos && bitpos+bitsize-1 <= bitregion_end

Presumably to satisfy the consecutive bitfields are a single memorylocation stuff from C++11. Thus the bitregion representation (C++11)must totally encompass the original COMPONENT_REF. I can certainly seehow that holds immediately after the call to get_bit_range.

When we adjust the address, we're just accounting for the bit position(bitpos). ie, if bitpos was 16 and TO_RTX was (plus (reg) (32)). Wechange TO_RTX to (plus (reg) (48)). This is ultimately meant tosimplify the resulting insns we generate by folding the bitposadjustment into the address computation. In effect, this results inaccessing a subobject rather than the full object. Presumably this iskosher for C++11's memory model (I really don't know).

*If* it is OK to access the subobject like this in C++11's memory model,then is is necessary for bitregion_{start,end} to encompass the originalobject, or should it be made consistent with the bitpos/bitsize? Andthat seems to me to be the fundamental question. What is really thepurpose of the bitregion_{start,end} representation? If it mustrepresent the C++ object, then ISTM we don't want to mess with it. Ifit doesn't, then why did we bother building this alternaterepresentation to begin with?



Jeff

Follow-Ups:
- RE: [PE-POST] Adjust Bit-region in expand_assignment
  - From: Bernd Edlinger

References:
- [PE-POST] Adjust Bit-region in expand_assignment
  - From: Bernd Edlinger
- Re: [PE-POST] Adjust Bit-region in expand_assignment
  - From: Jeff Law
- RE: [PE-POST] Adjust Bit-region in expand_assignment
  - From: Bernd Edlinger

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