Re: [Xen-devel] [PATCH 1/4] xen/PVH: Replace GDT_ENTRY with explicit constant

[Boris Ostrovsky <boris.ostrovsky@oracle.com>]

On 05/02/2018 04:00 AM, Jan Beulich wrote:
>>>> On 30.04.18 at 18:23, <boris.ostrovsky@oracle.com> wrote:
>> Latest binutils release (2.29.1) will no longer allow proper computation
>> of GDT entries on 32-bits, with warning:
>>
>> arch/x86/xen/xen-pvh.S: Assembler messages:
>> arch/x86/xen/xen-pvh.S:150: Warning: shift count out of range (32 is not between 0 and 31)
>> arch/x86/xen/xen-pvh.S:150: Warning: shift count out of range (40 is not between 0 and 31)
>> arch/x86/xen/xen-pvh.S:150: Warning: shift count out of range (32 is not between 0 and 31)
>> arch/x86/xen/xen-pvh.S:152: Warning: shift count out of range (32 is not between 0 and 31)
>> arch/x86/xen/xen-pvh.S:152: Warning: shift count out of range (40 is not between 0 and 31)
>> arch/x86/xen/xen-pvh.S:152: Warning: shift count out of range (32 is not between 0 and 31)
> I think this is a mis-configured binutils build - even if targeting 32-bit only, it
> should allow 64-bit arithmetic (i.e. be configured with --enable-64-bit-bfd).
> Note how, for example, this
>
> 	.long	1 << 32, 0x10000 * 0x10000
> 	.quad	1 << 32, 0x10000 * 0x10000
>
> assembles consistently with a warning on _both_ values on the first line
> with what I'd call a properly configured binutils build, but errors only on
> the shift expressions on each line for an (imo) improperly configured one.
> The only viable alternative would imo be to simply disallow .quad without
> --enable-64-bit-bfd, but I guess that would break a number of consumers.
>
> In any event I'd like to suggest to drop this patch.


Let me see go back and see how I built my binutils. And maybe rebuild
them on something > fedora13.

-boris