Re: git: 6713be315900 - main - Add NT_ARM_ADDR_MASK

On 2/22/22 9:59 AM, Jessica Clarke wrote:
> On 22 Feb 2022, at 17:27, Andrew Turner <andrew@FreeBSD.org> wrote:
>>
>> The branch main has been updated by andrew:
>>
>> URL: https://cgit.FreeBSD.org/src/commit/?id=6713be3159000783f7aacf3ea90d6c6878c44da0
>>
>> commit 6713be3159000783f7aacf3ea90d6c6878c44da0
>> Author:     Andrew Turner <andrew@FreeBSD.org>
>> AuthorDate: 2021-08-30 16:43:22 +0000
>> Commit:     Andrew Turner <andrew@FreeBSD.org>
>> CommitDate: 2022-02-22 17:10:35 +0000
>>
>>     Add NT_ARM_ADDR_MASK
>>
>>     This can be used by debuggers to find which bits in a virtual address
>>     should be masked off to get a canonical address. This is currently used
>>     by the Pointer Authentication Code support to get its mask. It could also
>>     be used if we support Top Byte Ignore for the same purpose.
>>
>>     Reviewed by:    kib
>>     Sponsored by:   The FreeBSD Foundation
>>     Differential Revision: https://reviews.freebsd.org/D34302
> 
> This definition seems backwards? In Linux it’s NT_ARM_PAC_MASK, i.e.
> the mask for the PAC bits, but you’ve called it NT_ARM_ADDR_MASK here,
> which sounds like the mask you apply to get the address bits, but it’s
> not, you apply the negation of this mask to get the address bits.

In that sense, it's identical to NT_ARM_PAC_MASK on Linux (same actual
contents and layout).  Given that, using a different name seems gratuitous
vs reusing the existing name as we did for NT_X86_XSTATE, NT_ARM_VFP,
NT_PPC_VMX, NT_ARM_TLS, etc.  If the content of the note was different,
then a different name might be warranted, but if the content and purpose
of the note are the same, then it's not clear why we need a different name
for the constant.

-- 
John Baldwin