From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from foss.arm.com (foss.arm.com [217.140.110.172])
 by mx.groups.io with SMTP id smtpd.web11.10331.1590079544865171079
 for <devel@edk2.groups.io>;
 Thu, 21 May 2020 09:45:45 -0700
Authentication-Results: mx.groups.io;
 dkim=missing; spf=pass (domain: arm.com, ip: 217.140.110.172, mailfrom: ard.biesheuvel@arm.com)
Received: from usa-sjc-imap-foss1.foss.arm.com (unknown [10.121.207.14])
	by usa-sjc-mx-foss1.foss.arm.com (Postfix) with ESMTP id 8C90AD6E;
	Thu, 21 May 2020 09:45:43 -0700 (PDT)
Received: from [192.168.1.81] (unknown [10.37.8.250])
	by usa-sjc-imap-foss1.foss.arm.com (Postfix) with ESMTPSA id 1D9DB3F68F;
	Thu, 21 May 2020 09:45:41 -0700 (PDT)
Subject: Re: [edk2-devel] [PATCH v2] ArmPkg/CompilerIntrinsicsLib: provide atomics intrinsics
To: devel@edk2.groups.io, philmd@redhat.com
Cc: glin@suse.com, leif@nuviainc.com, lersek@redhat.com, liming.gao@intel.com
References: <20200520114448.26104-1-ard.biesheuvel@arm.com>
 <47f54425-df5d-17a3-e134-fe9e01fb08bd@redhat.com>
 <6c9ec3a3-8aa1-c4d4-c7ba-1b9e28fd0866@redhat.com>
From: "Ard Biesheuvel" <ard.biesheuvel@arm.com>
Message-ID: <c5ade5be-9c21-9797-df72-c42317445b4d@arm.com>
Date: Thu, 21 May 2020 18:45:39 +0200
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:68.0) Gecko/20100101
 Thunderbird/68.7.0
MIME-Version: 1.0
In-Reply-To: <6c9ec3a3-8aa1-c4d4-c7ba-1b9e28fd0866@redhat.com>
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Language: en-US
Content-Transfer-Encoding: quoted-printable

On 5/21/20 6:40 PM, Philippe Mathieu-Daud=C3=A9 via groups.io wrote:
> On 5/20/20 2:37 PM, Philippe Mathieu-Daud=C3=A9 wrote:
>> Hi Ard,
>>
>> On 5/20/20 1:44 PM, Ard Biesheuvel wrote:
>>> Gary reports the GCC 10 will emit calls to atomics intrinsics routines
>>> unless -mno-outline-atomics is specified. This means GCC-10 introduces
>>> new intrinsics, and even though it would be possible to work around th=
is
>>> by specifying the command line option, this would require a new GCC10
>>> toolchain profile to be created, which we prefer to avoid.
>>>
>>> So instead, add the new intrinsics to our library so they are provided
>>> when necessary.
>>>
>>> Signed-off-by: Ard Biesheuvel <ard.biesheuvel@arm.com>
>>> ---
>>> v2:
>>> - add missing .globl to export the functions from the object file
>>> - add function end markers so the size of each is visible in the ELF=
=20
>>> metadata
>>> - add some comments to describe what is going on
>>
>> Thanks, head hurts a bit less...
>>
>>>
>>> =C2=A0 ArmPkg/Library/CompilerIntrinsicsLib/CompilerIntrinsicsLib.inf =
|=C2=A0=C2=A0 3 +
>>> =C2=A0 ArmPkg/Library/CompilerIntrinsicsLib/AArch64/Atomics.S=C2=A0=C2=
=
=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 |=20
>>> 142 ++++++++++++++++++++
>>> =C2=A0 2 files changed, 145 insertions(+)
>>>
>>> diff --git=20
>>> a/ArmPkg/Library/CompilerIntrinsicsLib/CompilerIntrinsicsLib.inf=20
>>> b/ArmPkg/Library/CompilerIntrinsicsLib/CompilerIntrinsicsLib.inf
>>> index d5bad9467758..fcf48c678119 100644
>>> --- a/ArmPkg/Library/CompilerIntrinsicsLib/CompilerIntrinsicsLib.inf
>>> +++ b/ArmPkg/Library/CompilerIntrinsicsLib/CompilerIntrinsicsLib.inf
>>> @@ -79,6 +79,9 @@ [Sources.ARM]
>>> =C2=A0=C2=A0=C2=A0 Arm/ldivmod.asm=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 | MSF=
T
>>> =C2=A0=C2=A0=C2=A0 Arm/llsr.asm=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=
=A0=C2=A0 | MSFT
>>> +[Sources.AARCH64]
>>> +=C2=A0 AArch64/Atomics.S=C2=A0=C2=A0=C2=A0 | GCC
>>> +
>>> =C2=A0 [Packages]
>>> =C2=A0=C2=A0=C2=A0 MdePkg/MdePkg.dec
>>> =C2=A0=C2=A0=C2=A0 ArmPkg/ArmPkg.dec
>>> diff --git a/ArmPkg/Library/CompilerIntrinsicsLib/AArch64/Atomics.S=20
>>> b/ArmPkg/Library/CompilerIntrinsicsLib/AArch64/Atomics.S
>>> new file mode 100644
>>> index 000000000000..dc61d6bb8e52
>>> --- /dev/null
>>> +++ b/ArmPkg/Library/CompilerIntrinsicsLib/AArch64/Atomics.S
>>> @@ -0,0 +1,142 @@
>>> +#--------------------------------------------------------------------=
----------=20
>>>
>>> +#
>>> +# Copyright (c) 2020, Arm, Limited. All rights reserved.<BR>
>>> +#
>>> +# SPDX-License-Identifier: BSD-2-Clause-Patent
>>> +#
>>> +#--------------------------------------------------------------------=
----------=20
>>>
>>> +
>>> +=C2=A0=C2=A0=C2=A0 /*
>>> +=C2=A0=C2=A0=C2=A0=C2=A0 * Provide the GCC intrinsics that are requir=
ed when using GCC 9 or
>>> +=C2=A0=C2=A0=C2=A0=C2=A0 * later with the -moutline-atomics options (=
which became the=20
>>> default
>>> +=C2=A0=C2=A0=C2=A0=C2=A0 * in GCC 10)
>>> +=C2=A0=C2=A0=C2=A0=C2=A0 */
>>> +=C2=A0=C2=A0=C2=A0 .arch armv8-a
>>> +
>>> +=C2=A0=C2=A0=C2=A0 .macro=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 r=
eg_alias, pfx, sz
>>> +=C2=A0=C2=A0=C2=A0 r0_\sz=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 .=
req=C2=A0=C2=A0=C2=A0 \pfx\()0
>>> +=C2=A0=C2=A0=C2=A0 r1_\sz=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 .=
req=C2=A0=C2=A0=C2=A0 \pfx\()1
>>> +=C2=A0=C2=A0=C2=A0 tmp0_\sz=C2=A0=C2=A0=C2=A0 .req=C2=A0=C2=A0=C2=A0 =
\pfx\()16
>>> +=C2=A0=C2=A0=C2=A0 tmp1_\sz=C2=A0=C2=A0=C2=A0 .req=C2=A0=C2=A0=C2=A0 =
\pfx\()17
>>> +=C2=A0=C2=A0=C2=A0 .endm
>>> +
>>> +=C2=A0=C2=A0=C2=A0 /*
>>> +=C2=A0=C2=A0=C2=A0=C2=A0 * Define register aliases of the right type =
for each size
>>> +=C2=A0=C2=A0=C2=A0=C2=A0 * (xN for 8 bytes, wN for everything smaller=
)
>>> +=C2=A0=C2=A0=C2=A0=C2=A0 */
>>> +=C2=A0=C2=A0=C2=A0 reg_alias=C2=A0=C2=A0=C2=A0 w, 1
>>> +=C2=A0=C2=A0=C2=A0 reg_alias=C2=A0=C2=A0=C2=A0 w, 2
>>> +=C2=A0=C2=A0=C2=A0 reg_alias=C2=A0=C2=A0=C2=A0 w, 4
>>> +=C2=A0=C2=A0=C2=A0 reg_alias=C2=A0=C2=A0=C2=A0 x, 8
>>> +
>>> +=C2=A0=C2=A0=C2=A0 .macro=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 f=
n_start, name:req
>>> +=C2=A0=C2=A0=C2=A0 .section=C2=A0=C2=A0=C2=A0 .text.\name
>>> +=C2=A0=C2=A0=C2=A0 .globl=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 \=
name
>>> +=C2=A0=C2=A0=C2=A0 .type=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 \n=
ame, %function
>>> +\name\():
>>> +=C2=A0=C2=A0=C2=A0 .endm
>>> +
>>> +=C2=A0=C2=A0=C2=A0 .macro=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 f=
n_end, name:req
>>> +=C2=A0=C2=A0=C2=A0 .size=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 \n=
ame, . - \name
>>> +=C2=A0=C2=A0=C2=A0 .endm
>>> +
>>> +=C2=A0=C2=A0=C2=A0 /*
>>> +=C2=A0=C2=A0=C2=A0=C2=A0 * Emit an atomic helper for \model with oper=
ands of size \sz,=20
>>> using
>>> +=C2=A0=C2=A0=C2=A0=C2=A0 * the operation specified by \insn (which is=
 the LSE name), and=20
>>> which
>>> +=C2=A0=C2=A0=C2=A0=C2=A0 * can be implemented using the generic=20
>>> load-locked/store-conditional
>>> +=C2=A0=C2=A0=C2=A0=C2=A0 * (LL/SC) sequence below, using the arithmet=
ic operation given by
>>> +=C2=A0=C2=A0=C2=A0=C2=A0 * \opc.
>>> +=C2=A0=C2=A0=C2=A0=C2=A0 */
>>> +=C2=A0=C2=A0=C2=A0 .macro=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0 emit_ld_sz, sz:req, insn:req, opc:req, model:req,=20
>>> s, a, l
>>> +=C2=A0=C2=A0=C2=A0 fn_start=C2=A0=C2=A0=C2=A0 __aarch64_\insn\()\sz\(=
)\model
>>> +=C2=A0=C2=A0=C2=A0 mov=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 tmp0=
_\sz, r0_\sz
>>> +0:=C2=A0=C2=A0=C2=A0 ld\a\()xr\s=C2=A0=C2=A0=C2=A0 r0_\sz, [x1]
>>> +=C2=A0=C2=A0=C2=A0 .ifnc=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 \i=
nsn, swp
>>> +=C2=A0=C2=A0=C2=A0 \opc=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 tmp=
1_\sz, r0_\sz, tmp0_\sz
>>> +=C2=A0=C2=A0=C2=A0 .else
>>> +=C2=A0=C2=A0=C2=A0 \opc=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 tmp=
1_\sz, tmp0_\sz
>>> +=C2=A0=C2=A0=C2=A0 .endif
>>> +=C2=A0=C2=A0=C2=A0 st\l\()xr\s=C2=A0=C2=A0=C2=A0 w15, tmp1_\sz, [x1]
>>> +=C2=A0=C2=A0=C2=A0 cbnz=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 w15=
, 0b
>>
>> I see at the end \s is in {,b,h} range.
>>
>> Don't you need to use x15 on 64-bit?
>=20
> Ard, I expanded all macros and reviewed this patch, but I am still=20
> having hard time to figure why w15 temp is OK instead of x15. Any hint?
>=20

Why do you think it should be x15?


>>
>>> +=C2=A0=C2=A0=C2=A0 ret
>>> +=C2=A0=C2=A0=C2=A0 fn_end=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 _=
_aarch64_\insn\()\sz\()\model
>>> +=C2=A0=C2=A0=C2=A0 .endm
>>> +
>>> +=C2=A0=C2=A0=C2=A0 /*
>>> +=C2=A0=C2=A0=C2=A0=C2=A0 * Emit atomic helpers for \model for operand=
 sizes in the
>>> +=C2=A0=C2=A0=C2=A0=C2=A0 * set {1, 2, 4, 8}, for the instruction patt=
ern given by
>>> +=C2=A0=C2=A0=C2=A0=C2=A0 * \insn. (This is the LSE name, but this imp=
lementation uses
>>> +=C2=A0=C2=A0=C2=A0=C2=A0 * the generic LL/SC sequence using \opc as t=
he arithmetic
>>> +=C2=A0=C2=A0=C2=A0=C2=A0 * operation on the target.)
>>> +=C2=A0=C2=A0=C2=A0=C2=A0 */
>>> +=C2=A0=C2=A0=C2=A0 .macro=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 e=
mit_ld, insn:req, opc:req, model:req, a, l
>>> +=C2=A0=C2=A0=C2=A0 emit_ld_sz=C2=A0=C2=A0=C2=A0 1, \insn, \opc, \mode=
l, b, \a, \l
>>> +=C2=A0=C2=A0=C2=A0 emit_ld_sz=C2=A0=C2=A0=C2=A0 2, \insn, \opc, \mode=
l, h, \a, \l
>>> +=C2=A0=C2=A0=C2=A0 emit_ld_sz=C2=A0=C2=A0=C2=A0 4, \insn, \opc, \mode=
l,=C2=A0 , \a, \l
>>> +=C2=A0=C2=A0=C2=A0 emit_ld_sz=C2=A0=C2=A0=C2=A0 8, \insn, \opc, \mode=
l,=C2=A0 , \a, \l
>>> +=C2=A0=C2=A0=C2=A0 .endm
>>> +
>>> +=C2=A0=C2=A0=C2=A0 /*
>>> +=C2=A0=C2=A0=C2=A0=C2=A0 * Emit the compare and swap helper for \mode=
l and size \sz
>>> +=C2=A0=C2=A0=C2=A0=C2=A0 * using LL/SC instructions.
>>> +=C2=A0=C2=A0=C2=A0=C2=A0 */
>>> +=C2=A0=C2=A0=C2=A0 .macro=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
=C2=A0 emit_cas_sz, sz:req, model:req, uxt:req, s, a, l
>>> +=C2=A0=C2=A0=C2=A0 fn_start=C2=A0=C2=A0=C2=A0 __aarch64_cas\sz\()\mod=
el
>>> +=C2=A0=C2=A0=C2=A0 \uxt=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 tmp=
0_\sz, r0_\sz
>>> +0:=C2=A0=C2=A0=C2=A0 ld\a\()xr\s=C2=A0=C2=A0=C2=A0 r0_\sz, [x2]
>>> +=C2=A0=C2=A0=C2=A0 cmp=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 r0_\=
sz, tmp0_\sz
>>> +=C2=A0=C2=A0=C2=A0 bne=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 1f
>>> +=C2=A0=C2=A0=C2=A0 st\l\()xr\s=C2=A0=C2=A0=C2=A0 w15, r1_\sz, [x2]
>>> +=C2=A0=C2=A0=C2=A0 cbnz=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 w15=
, 0b
>>> +1:=C2=A0=C2=A0=C2=A0 ret
>>> +=C2=A0=C2=A0=C2=A0 fn_end=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 _=
_aarch64_cas\sz\()\model
>>> +=C2=A0=C2=A0=C2=A0 .endm
>>> +
>>> +=C2=A0=C2=A0=C2=A0 /*
>>> +=C2=A0=C2=A0=C2=A0=C2=A0 * Emit compare-and-swap helpers for \model f=
or operand sizes in=20
>>> the
>>> +=C2=A0=C2=A0=C2=A0=C2=A0 * set {1, 2, 4, 8, 16}.
>>> +=C2=A0=C2=A0=C2=A0=C2=A0 */
>>> +=C2=A0=C2=A0=C2=A0 .macro=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 e=
mit_cas, model:req, a, l
>>> +=C2=A0=C2=A0=C2=A0 emit_cas_sz=C2=A0=C2=A0=C2=A0 1, \model, uxtb, b, =
\a, \l
>>> +=C2=A0=C2=A0=C2=A0 emit_cas_sz=C2=A0=C2=A0=C2=A0 2, \model, uxth, h, =
\a, \l
>>> +=C2=A0=C2=A0=C2=A0 emit_cas_sz=C2=A0=C2=A0=C2=A0 4, \model, mov ,=C2=
=A0 , \a, \l
>>> +=C2=A0=C2=A0=C2=A0 emit_cas_sz=C2=A0=C2=A0=C2=A0 8, \model, mov ,=C2=
=A0 , \a, \l
>>> +
>>> +=C2=A0=C2=A0=C2=A0 /*
>>> +=C2=A0=C2=A0=C2=A0=C2=A0 * We cannot use the parameterized sequence f=
or 16 byte CAS, so we
>>> +=C2=A0=C2=A0=C2=A0=C2=A0 * need to define it explicitly.
>>> +=C2=A0=C2=A0=C2=A0=C2=A0 */
>>> +=C2=A0=C2=A0=C2=A0 fn_start=C2=A0=C2=A0=C2=A0 __aarch64_cas16\model
>>> +=C2=A0=C2=A0=C2=A0 mov=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 x16,=
 x0
>>> +=C2=A0=C2=A0=C2=A0 mov=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 x17,=
 x1
>>> +0:=C2=A0=C2=A0=C2=A0 ld\a\()xp=C2=A0=C2=A0=C2=A0 x0, x1, [x4]
>>> +=C2=A0=C2=A0=C2=A0 cmp=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 x0, =
x16
>>> +=C2=A0=C2=A0=C2=A0 ccmp=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 x1,=
 x17, #0, eq
>>> +=C2=A0=C2=A0=C2=A0 bne=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 1f
>>> +=C2=A0=C2=A0=C2=A0 st\l\()xp=C2=A0=C2=A0=C2=A0 w15, x16, x17, [x4]
>>> +=C2=A0=C2=A0=C2=A0 cbnz=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 w15=
, 0b
>>> +1:=C2=A0=C2=A0=C2=A0 ret
>>> +=C2=A0=C2=A0=C2=A0 fn_end=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 _=
_aarch64_cas16\model
>>> +=C2=A0=C2=A0=C2=A0 .endm
>>> +
>>> +=C2=A0=C2=A0=C2=A0 /*
>>> +=C2=A0=C2=A0=C2=A0=C2=A0 * Emit the set of GCC outline atomic helper =
functions for
>>> +=C2=A0=C2=A0=C2=A0=C2=A0 * the memory ordering model given by \model:
>>> +=C2=A0=C2=A0=C2=A0=C2=A0 * - relax=C2=A0=C2=A0=C2=A0 unordered loads =
and stores
>>> +=C2=A0=C2=A0=C2=A0=C2=A0 * - acq=C2=A0=C2=A0=C2=A0 load-acquire, unor=
dered store
>>> +=C2=A0=C2=A0=C2=A0=C2=A0 * - rel=C2=A0=C2=A0=C2=A0 unordered load, st=
ore-release
>>> +=C2=A0=C2=A0=C2=A0=C2=A0 * - acq_rel=C2=A0=C2=A0=C2=A0 load-acquire, =
store-release
>>> +=C2=A0=C2=A0=C2=A0=C2=A0 */
>>> +=C2=A0=C2=A0=C2=A0 .macro=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 e=
mit_model, model:req, a, l
>>> +=C2=A0=C2=A0=C2=A0 emit_ld=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 =
ldadd, add, \model, \a, \l
>>> +=C2=A0=C2=A0=C2=A0 emit_ld=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 =
ldclr, bic, \model, \a, \l
>>> +=C2=A0=C2=A0=C2=A0 emit_ld=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 =
ldeor, eor, \model, \a, \l
>>> +=C2=A0=C2=A0=C2=A0 emit_ld=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 =
ldset, orr, \model, \a, \l
>>> +=C2=A0=C2=A0=C2=A0 emit_ld=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 =
swp,=C2=A0=C2=A0 mov, \model, \a, \l
>>> +=C2=A0=C2=A0=C2=A0 emit_cas=C2=A0=C2=A0=C2=A0 \model, \a, \l
>>> +=C2=A0=C2=A0=C2=A0 .endm
>>> +
>>> +=C2=A0=C2=A0=C2=A0 emit_model=C2=A0=C2=A0=C2=A0 _relax
>>> +=C2=A0=C2=A0=C2=A0 emit_model=C2=A0=C2=A0=C2=A0 _acq, a
>>> +=C2=A0=C2=A0=C2=A0 emit_model=C2=A0=C2=A0=C2=A0 _rel,, l
>>> +=C2=A0=C2=A0=C2=A0 emit_model=C2=A0=C2=A0=C2=A0 _acq_rel, a, l
>>>
>=20
>=20
>=20
>=20