XCore target, getPreferredTypeAlign() *MUST* return preferred alignment.

[Robert Lytton]

Thanks Rafael,
Will amend the comment.
Robert
________________________________________
From: Rafael Espíndola [rafael.espindola at gmail.com]
Sent: 01 November 2013 19:46
To: Robert Lytton
Cc: cfe-commits at cs.uiuc.edu
Subject: Re: XCore target, getPreferredTypeAlign() *MUST* return preferred alignment.

The comment looks wrong. I assume you want something like "never
overalign on xcore", no?

I assume the change is correct from xcore's point of view, so LGTM
with the comment fixed.

On 1 November 2013 13:16, Robert Lytton <robert at xmos.com> wrote:
> Ping!
>
> ________________________________
> From: Robert Lytton
> Sent: 21 October 2013 10:17
>
> To: cfe-commits at cs.uiuc.edu
> Subject: RE: XCore target, getPreferredTypeAlign() *MUST* return preferred
> alignment.
>
> Hi
> Is this good to go?
> Thank you.
> Robert
> ________________________________
> From: cfe-commits-bounces at cs.uiuc.edu [cfe-commits-bounces at cs.uiuc.edu] on
> behalf of Robert Lytton [robert at xmos.com]
> Sent: 16 October 2013 17:03
> To: cfe-commits at cs.uiuc.edu
> Subject: RE: XCore target, getPreferredTypeAlign() *MUST* return preferred
> alignment.
>
> Hi
>
> I've moved the _Static_assert tests from test/Driver/xcore-opts.c into
> test/CodeGen/xcore-abi.c.
> Is the patch good to go?
>
> Robert
>
> ________________________________
> From: cfe-commits-bounces at cs.uiuc.edu [cfe-commits-bounces at cs.uiuc.edu] on
> behalf of Robert Lytton [robert at xmos.com]
> Sent: 14 October 2013 13:33
> To: cfe-commits at cs.uiuc.edu
> Subject: XCore target, getPreferredTypeAlign() *MUST* return preferred
> alignment.
>
> Hi
>
> Here is a suggest patch to support the XCore target.
>
> This is required as the the xcore llvm backend does not handle 8 byte
> alignment viz:
>         %BadAlignment = alloca i64, align 8
>
> (llvm will reject the above as an error)
>
>
> The patch prevents the default behavior running:
>     /// getPreferredTypeAlign - Return the "preferred" alignment of the
> specified
>     /// type for the current target in bits.  This can be different than the
> ABI
>     /// alignment in cases where it is beneficial for performance to
> overalign
>     /// a data type.
>
>
> Robert
>
>
> _______________________________________________
> cfe-commits mailing list
> cfe-commits at cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
>