http://gcc.gnu.org/bugzilla/show_bug.cgi?id=42652
Richard Guenther rguenth at gcc dot gnu.org changed:
What|Removed |Added
Blocks||53947
--- Comment #17 from irar at il dot ibm dot com 2010-02-22 09:01 ---
Is there a way to pass alignment information similar to PR 39954?
Otherwise, a proper fix would be some inter-procedural analysis... Meantime, we
can do intra-procedural analysis and fail when we reach function
--- Comment #18 from rguenth at gcc dot gnu dot org 2010-02-22 10:27
---
(In reply to comment #17)
Is there a way to pass alignment information similar to PR 39954?
Well, that wasn't really a true fix but simply restored the (broken) behavior
of earlier compilers so the 4.5
--- Comment #14 from law at redhat dot com 2010-02-09 23:04 ---
Subject: Re: vectorizer created unaligned vector
insns
On 01/18/10 05:17, irar at il dot ibm dot com wrote:
--- Comment #13 from irar at il dot ibm dot com 2010-01-18 12:17 ---
Does something like this make
--- Comment #15 from rguenth at gcc dot gnu dot org 2010-02-09 23:11
---
(In reply to comment #14)
Subject: Re: vectorizer created unaligned vector
insns
On 01/18/10 05:17, irar at il dot ibm dot com wrote:
--- Comment #13 from irar at il dot ibm dot com 2010-01-18 12:17
--- Comment #16 from law at redhat dot com 2010-02-09 23:49 ---
Subject: Re: vectorizer created unaligned vector
insns
On 02/09/10 16:11, rguenth at gcc dot gnu dot org wrote:
--- Comment #15 from rguenth at gcc dot gnu dot org 2010-02-09 23:11
---
(In reply to comment
--- Comment #13 from irar at il dot ibm dot com 2010-01-18 12:17 ---
Does something like this make sense? (With this patch we will never use peeling
for function parameters, unless the builtin returns OK to peel for packed
types).
Index: tree-vect-data-refs.c
--- Comment #11 from law at redhat dot com 2010-01-14 17:29 ---
Subject: Re: vectorizer created unaligned vector
insns
On 01/13/10 02:35, irar at il dot ibm dot com wrote:
--- Comment #10 from irar at il dot ibm dot com 2010-01-13 09:35 ---
Yes, I understand that we can't
--- Comment #12 from hjl dot tools at gmail dot com 2010-01-14 17:34
---
Intel AVX architecture won't fault on unaligned load-op instructions
and unaligned VMOV* instructions are as fast as aligned VMOV*
instructions on aligned memory. For AVX, we can always use unaligned
VMOV*
--- Comment #10 from irar at il dot ibm dot com 2010-01-13 09:35 ---
Yes, I understand that we can't assume that an access is aligned if we can't
prove it's aligned. I don't understand how we can prove that a COMPONENT_REF is
aligned, i.e., if there is a way to check if a struct is
--- Comment #8 from irar at il dot ibm dot com 2010-01-12 08:08 ---
So, to be on the safe side, we should assume that COMPONENT_REFs are not
naturally aligned and never use peeling for them?
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=42652
--- Comment #9 from law at redhat dot com 2010-01-12 15:18 ---
Subject: Re: vectorizer created unaligned vector
insns
On 01/12/10 01:08, irar at il dot ibm dot com wrote:
--- Comment #8 from irar at il dot ibm dot com 2010-01-12 08:08 ---
So, to be on the safe side, we
--- Comment #6 from law at redhat dot com 2010-01-11 17:14 ---
Subject: Re: vectorizer created unaligned vector
insns
On 01/10/10 01:22, irar at il dot ibm dot com wrote:
--- Comment #5 from irar at il dot ibm dot com 2010-01-10 08:22 ---
In vector_alignment_reachable_p()
--- Comment #7 from law at redhat dot com 2010-01-11 17:16 ---
Subject: Re: vectorizer created unaligned vector
insns
On 01/08/10 10:25, rguenth at gcc dot gnu dot org wrote:
--- Comment #4 from rguenth at gcc dot gnu dot org 2010-01-08 17:25
---
Well, indeed we have a
--- Comment #5 from irar at il dot ibm dot com 2010-01-10 08:22 ---
In vector_alignment_reachable_p() we check if an access is packed using
contains_packed_reference(). For packed accesses we return false, meaning
alignment is unreachable and peeling cannot be used.
In the attached
--- Comment #2 from rguenth at gcc dot gnu dot org 2010-01-08 11:58 ---
There is some hacks with contains_packed_reference but IIRC I was complaining
about these at some point - they are not conservatively correct.
--
rguenth at gcc dot gnu dot org changed:
What
--- Comment #3 from law at redhat dot com 2010-01-08 16:45 ---
Subject: Re: vectorizer created unaligned vector
insns
On 01/08/10 04:58, rguenth at gcc dot gnu dot org wrote:
--- Comment #2 from rguenth at gcc dot gnu dot org 2010-01-08 11:58
---
There is some hacks with
--- Comment #4 from rguenth at gcc dot gnu dot org 2010-01-08 17:25 ---
Well, indeed we have a certain weakness in how we represent pointers to
possibly (un-)aligned stuff. See PR39954 for another case. Maybe this
bug is really a duplicate of the underlying problems.
--
--- Comment #1 from law at redhat dot com 2010-01-07 17:33 ---
Created an attachment (id=19501)
-- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=19501action=view)
testcase
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=42652
19 matches
Mail list logo
