* vector alignment
@ 2019-04-02 16:20 Martin Sebor
2019-04-03 11:14 ` Richard Biener
0 siblings, 1 reply; 6+ messages in thread
From: Martin Sebor @ 2019-04-02 16:20 UTC (permalink / raw)
To: gcc mailing list
GCC tries to align a vector on its natural boundary, i.e., that
given by its size, up to MAX_OBJECT_ALIGNMENT. Vectors that are
bigger than that are either silently [mis]aligned on that same
maximum boundary (PR 89798), silently truncated (and misaligned),
or cause an ICE (PR 89797). Compiling the following:
__attribute__ ((vector_size (N))) char v;
_Static_assert (sizeof (v) == N, "size");
_Static_assert (__alignof__ (v) == N, "alignment");
with N set to 1LLU << I shows these failures:
I < 29 succeeds
I < 31 fails alignment
I < 32 ICE
I >= 32 fails alignment and size
Attribute aligned doesn't seem to have any effect on types or
variables declared with attribute vector_size. The alignment
set by the latter prevails.
This happens no matter what scope the vector is defined in (i.e.,
file or local).
I have some questions:
1) Is there some reason to align vectors on the same boundary
as their size no matter how big it is? I can't find such
a requirement in the ABIs I looked at. Or would it be more
appropriate to align the big ones on the preferred boundary
for the target? For instance, does it make more sense to
align a 64KB vector on a 64KB boundary than on, say,
a 64-byte boundary (or some other boundary less than 64K?)
2) If not, is it then appropriate to underalign very large
vectors on a boundary less than their size?
3) Should the aligned attribute not override the default vector
alignment?
I would like to think the answer to (1) is that vectors should
be aligned on the preferred boundary for the target/ABI. If
that's feasible, it should also obviate question (2).
I believe the answer to (3) is yes. If not, GCC should issue
a warning that it doesn't honor the aligned attribute.
Thanks
Martin
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: vector alignment
2019-04-02 16:20 vector alignment Martin Sebor
@ 2019-04-03 11:14 ` Richard Biener
2019-04-03 17:59 ` Martin Sebor
2019-04-17 15:12 ` Joseph Myers
0 siblings, 2 replies; 6+ messages in thread
From: Richard Biener @ 2019-04-03 11:14 UTC (permalink / raw)
To: Martin Sebor; +Cc: gcc mailing list
On Tue, Apr 2, 2019 at 6:20 PM Martin Sebor <msebor@gmail.com> wrote:
>
> GCC tries to align a vector on its natural boundary, i.e., that
> given by its size, up to MAX_OBJECT_ALIGNMENT. Vectors that are
> bigger than that are either silently [mis]aligned on that same
> maximum boundary (PR 89798), silently truncated (and misaligned),
> or cause an ICE (PR 89797). Compiling the following:
>
> __attribute__ ((vector_size (N))) char v;
>
> _Static_assert (sizeof (v) == N, "size");
> _Static_assert (__alignof__ (v) == N, "alignment");
>
> with N set to 1LLU << I shows these failures:
>
> I < 29 succeeds
> I < 31 fails alignment
> I < 32 ICE
> I >= 32 fails alignment and size
>
> Attribute aligned doesn't seem to have any effect on types or
> variables declared with attribute vector_size. The alignment
> set by the latter prevails.
>
> This happens no matter what scope the vector is defined in (i.e.,
> file or local).
>
> I have some questions:
>
> 1) Is there some reason to align vectors on the same boundary
> as their size no matter how big it is? I can't find such
> a requirement in the ABIs I looked at. Or would it be more
> appropriate to align the big ones on the preferred boundary
> for the target? For instance, does it make more sense to
> align a 64KB vector on a 64KB boundary than on, say,
> a 64-byte boundary (or some other boundary less than 64K?)
I don't think there's a good reason. Instead I think that
BIGGEST_ALIGNMENT is what we should go for as upper limit,
anything bigger doesn't make sense (unless the user explicitely
requests it).
> 2) If not, is it then appropriate to underalign very large
> vectors on a boundary less than their size?
Yes.
> 3) Should the aligned attribute not override the default vector
> alignment?
Yes, but doesn't it already?
> I would like to think the answer to (1) is that vectors should
> be aligned on the preferred boundary for the target/ABI. If
> that's feasible, it should also obviate question (2).
>
> I believe the answer to (3) is yes. If not, GCC should issue
> a warning that it doesn't honor the aligned attribute.
>
> Thanks
> Martin
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: vector alignment
2019-04-03 11:14 ` Richard Biener
@ 2019-04-03 17:59 ` Martin Sebor
2019-04-04 5:40 ` Richard Biener
2019-04-17 15:12 ` Joseph Myers
1 sibling, 1 reply; 6+ messages in thread
From: Martin Sebor @ 2019-04-03 17:59 UTC (permalink / raw)
To: Richard Biener; +Cc: gcc mailing list
On 4/3/19 5:13 AM, Richard Biener wrote:
> On Tue, Apr 2, 2019 at 6:20 PM Martin Sebor <msebor@gmail.com> wrote:
>>
>> GCC tries to align a vector on its natural boundary, i.e., that
>> given by its size, up to MAX_OBJECT_ALIGNMENT. Vectors that are
>> bigger than that are either silently [mis]aligned on that same
>> maximum boundary (PR 89798), silently truncated (and misaligned),
>> or cause an ICE (PR 89797). Compiling the following:
>>
>> __attribute__ ((vector_size (N))) char v;
>>
>> _Static_assert (sizeof (v) == N, "size");
>> _Static_assert (__alignof__ (v) == N, "alignment");
>>
>> with N set to 1LLU << I shows these failures:
>>
>> I < 29 succeeds
>> I < 31 fails alignment
>> I < 32 ICE
>> I >= 32 fails alignment and size
>>
>> Attribute aligned doesn't seem to have any effect on types or
>> variables declared with attribute vector_size. The alignment
>> set by the latter prevails.
>>
>> This happens no matter what scope the vector is defined in (i.e.,
>> file or local).
>>
>> I have some questions:
>>
>> 1) Is there some reason to align vectors on the same boundary
>> as their size no matter how big it is? I can't find such
>> a requirement in the ABIs I looked at. Or would it be more
>> appropriate to align the big ones on the preferred boundary
>> for the target? For instance, does it make more sense to
>> align a 64KB vector on a 64KB boundary than on, say,
>> a 64-byte boundary (or some other boundary less than 64K?)
>
> I don't think there's a good reason. Instead I think that
> BIGGEST_ALIGNMENT is what we should go for as upper limit,
> anything bigger doesn't make sense (unless the user explicitely
> requests it).
Sounds good. Changing the alignment will impact object layout.
How do you suggest to deal with it? (Presumably for GCC 10.)
Issuing an ABI warning and adding an option to override
the new setting come to mind as possible mitigating solutions.
>
>> 2) If not, is it then appropriate to underalign very large
>> vectors on a boundary less than their size?
>
> Yes.
Ack.
>
>> 3) Should the aligned attribute not override the default vector
>> alignment?
>
> Yes, but doesn't it already?
Not if both are specified on the same declaration, as in:
typedef __attribute__ ((aligned (1024), vector_size (256))) int V;
I opened PR 89950 for this.
Martin
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: vector alignment
2019-04-03 17:59 ` Martin Sebor
@ 2019-04-04 5:40 ` Richard Biener
2019-04-04 16:14 ` Martin Sebor
0 siblings, 1 reply; 6+ messages in thread
From: Richard Biener @ 2019-04-04 5:40 UTC (permalink / raw)
To: Martin Sebor; +Cc: gcc mailing list
On April 3, 2019 7:59:47 PM GMT+02:00, Martin Sebor <msebor@gmail.com> wrote:
>On 4/3/19 5:13 AM, Richard Biener wrote:
>> On Tue, Apr 2, 2019 at 6:20 PM Martin Sebor <msebor@gmail.com> wrote:
>>>
>>> GCC tries to align a vector on its natural boundary, i.e., that
>>> given by its size, up to MAX_OBJECT_ALIGNMENT. Vectors that are
>>> bigger than that are either silently [mis]aligned on that same
>>> maximum boundary (PR 89798), silently truncated (and misaligned),
>>> or cause an ICE (PR 89797). Compiling the following:
>>>
>>> __attribute__ ((vector_size (N))) char v;
>>>
>>> _Static_assert (sizeof (v) == N, "size");
>>> _Static_assert (__alignof__ (v) == N, "alignment");
>>>
>>> with N set to 1LLU << I shows these failures:
>>>
>>> I < 29 succeeds
>>> I < 31 fails alignment
>>> I < 32 ICE
>>> I >= 32 fails alignment and size
>>>
>>> Attribute aligned doesn't seem to have any effect on types or
>>> variables declared with attribute vector_size. The alignment
>>> set by the latter prevails.
>>>
>>> This happens no matter what scope the vector is defined in (i.e.,
>>> file or local).
>>>
>>> I have some questions:
>>>
>>> 1) Is there some reason to align vectors on the same boundary
>>> as their size no matter how big it is? I can't find such
>>> a requirement in the ABIs I looked at. Or would it be more
>>> appropriate to align the big ones on the preferred boundary
>>> for the target? For instance, does it make more sense to
>>> align a 64KB vector on a 64KB boundary than on, say,
>>> a 64-byte boundary (or some other boundary less than 64K?)
>>
>> I don't think there's a good reason. Instead I think that
>> BIGGEST_ALIGNMENT is what we should go for as upper limit,
>> anything bigger doesn't make sense (unless the user explicitely
>> requests it).
>
>Sounds good. Changing the alignment will impact object layout.
Do we really apply the alignment there? How do other compilers lay out here?
>How do you suggest to deal with it? (Presumably for GCC 10.)
>Issuing an ABI warning and adding an option to override
>the new setting come to mind as possible mitigating solutions.
We could reject these vector types in aggregates in favor of arrays. Of course that ship has sailed...
>>
>>> 2) If not, is it then appropriate to underalign very large
>>> vectors on a boundary less than their size?
>>
>> Yes.
>
>Ack.
>
>>
>>> 3) Should the aligned attribute not override the default vector
>>> alignment?
>>
>> Yes, but doesn't it already?
>
>Not if both are specified on the same declaration, as in:
>
> typedef __attribute__ ((aligned (1024), vector_size (256))) int V;
>
>I opened PR 89950 for this.
>
>Martin
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: vector alignment
2019-04-04 5:40 ` Richard Biener
@ 2019-04-04 16:14 ` Martin Sebor
0 siblings, 0 replies; 6+ messages in thread
From: Martin Sebor @ 2019-04-04 16:14 UTC (permalink / raw)
To: Richard Biener; +Cc: gcc mailing list
On 4/3/19 11:40 PM, Richard Biener wrote:
> On April 3, 2019 7:59:47 PM GMT+02:00, Martin Sebor <msebor@gmail.com> wrote:
>> On 4/3/19 5:13 AM, Richard Biener wrote:
>>> On Tue, Apr 2, 2019 at 6:20 PM Martin Sebor <msebor@gmail.com> wrote:
>>>>
>>>> GCC tries to align a vector on its natural boundary, i.e., that
>>>> given by its size, up to MAX_OBJECT_ALIGNMENT. Vectors that are
>>>> bigger than that are either silently [mis]aligned on that same
>>>> maximum boundary (PR 89798), silently truncated (and misaligned),
>>>> or cause an ICE (PR 89797). Compiling the following:
>>>>
>>>> __attribute__ ((vector_size (N))) char v;
>>>>
>>>> _Static_assert (sizeof (v) == N, "size");
>>>> _Static_assert (__alignof__ (v) == N, "alignment");
>>>>
>>>> with N set to 1LLU << I shows these failures:
>>>>
>>>> I < 29 succeeds
>>>> I < 31 fails alignment
>>>> I < 32 ICE
>>>> I >= 32 fails alignment and size
>>>>
>>>> Attribute aligned doesn't seem to have any effect on types or
>>>> variables declared with attribute vector_size. The alignment
>>>> set by the latter prevails.
>>>>
>>>> This happens no matter what scope the vector is defined in (i.e.,
>>>> file or local).
>>>>
>>>> I have some questions:
>>>>
>>>> 1) Is there some reason to align vectors on the same boundary
>>>> as their size no matter how big it is? I can't find such
>>>> a requirement in the ABIs I looked at. Or would it be more
>>>> appropriate to align the big ones on the preferred boundary
>>>> for the target? For instance, does it make more sense to
>>>> align a 64KB vector on a 64KB boundary than on, say,
>>>> a 64-byte boundary (or some other boundary less than 64K?)
>>>
>>> I don't think there's a good reason. Instead I think that
>>> BIGGEST_ALIGNMENT is what we should go for as upper limit,
>>> anything bigger doesn't make sense (unless the user explicitely
>>> requests it).
>>
>> Sounds good. Changing the alignment will impact object layout.
>
> Do we really apply the alignment there? How do other compilers lay out here?
Unfortunately, yes. The struct below is 2048 in size.
struct S {
char c;
__attribute__ ((vector_size (1024))) char a;
};
Clang and ICC do the same thing. The biggest vector Clang allows
is 1024 bytes. ICC tops out at 1<<21. GCC ICEs for sizes over
1LLU << 32 but with my patch for bug 89797 it accepts any size
up to 1/2 the address space.
With types other than char the aligned attribute can be used in
conjunction with packed to reduce the size to just 1088:
struct S {
char c;
__attribute__ ((aligned (64), packed, vector_size (1024))) int a;
};
All three compilers require the packed attribute to reduce
the alignment. Clang and GCC silently ignore the aligned
attribute without it but ICC issues warning #1366: a reduction
in alignment without the "packed" attribute is ignored. That
would be a helpful enhancement to add to GCC.
I think ignoring the packed attribute for char vectors is a bug
so I opened PR 89968 for it.
>> How do you suggest to deal with it? (Presumably for GCC 10.)
>> Issuing an ABI warning and adding an option to override
>> the new setting come to mind as possible mitigating solutions.
>
> We could reject these vector types in aggregates in favor of arrays. Of course that ship has sailed...
We could still issue a warning for vectors that are excessively
overaligned and suggest to use attributes aligned and packed to
reduce the alignment.
Martin
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: vector alignment
2019-04-03 11:14 ` Richard Biener
2019-04-03 17:59 ` Martin Sebor
@ 2019-04-17 15:12 ` Joseph Myers
1 sibling, 0 replies; 6+ messages in thread
From: Joseph Myers @ 2019-04-17 15:12 UTC (permalink / raw)
To: Richard Biener; +Cc: Martin Sebor, gcc mailing list
On Wed, 3 Apr 2019, Richard Biener wrote:
> > 1) Is there some reason to align vectors on the same boundary
> > as their size no matter how big it is? I can't find such
> > a requirement in the ABIs I looked at. Or would it be more
> > appropriate to align the big ones on the preferred boundary
> > for the target? For instance, does it make more sense to
> > align a 64KB vector on a 64KB boundary than on, say,
> > a 64-byte boundary (or some other boundary less than 64K?)
>
> I don't think there's a good reason. Instead I think that
> BIGGEST_ALIGNMENT is what we should go for as upper limit,
> anything bigger doesn't make sense (unless the user explicitely
> requests it).
It's plausible that you'd want the ABI for a given vector type not to
depend on which architecture set extensions are enabled, but those
extensions might affect BIGGEST_ALIGNMENT (or the largest alignment
meaningful for any instruction, which is not necessarily the same as
BIGGEST_ALIGNMENT; Arm has some vector instructions that can use
alignments larger than BIGGEST_ALIGNMENT, for example).
--
Joseph S. Myers
joseph@codesourcery.com
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2019-04-17 15:12 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-04-02 16:20 vector alignment Martin Sebor
2019-04-03 11:14 ` Richard Biener
2019-04-03 17:59 ` Martin Sebor
2019-04-04 5:40 ` Richard Biener
2019-04-04 16:14 ` Martin Sebor
2019-04-17 15:12 ` Joseph Myers
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).