[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-devel] [PATCH v7 06/49] x86emul: support AVX512{F, BW, DQ} integer broadcast insns


  • To: Jan Beulich <JBeulich@xxxxxxxx>
  • From: Andrew Cooper <andrew.cooper3@xxxxxxxxxx>
  • Date: Fri, 15 Mar 2019 16:39:17 +0000
  • Autocrypt: addr=andrew.cooper3@xxxxxxxxxx; prefer-encrypt=mutual; keydata= mQINBFLhNn8BEADVhE+Hb8i0GV6mihnnr/uiQQdPF8kUoFzCOPXkf7jQ5sLYeJa0cQi6Penp VtiFYznTairnVsN5J+ujSTIb+OlMSJUWV4opS7WVNnxHbFTPYZVQ3erv7NKc2iVizCRZ2Kxn srM1oPXWRic8BIAdYOKOloF2300SL/bIpeD+x7h3w9B/qez7nOin5NzkxgFoaUeIal12pXSR Q354FKFoy6Vh96gc4VRqte3jw8mPuJQpfws+Pb+swvSf/i1q1+1I4jsRQQh2m6OTADHIqg2E ofTYAEh7R5HfPx0EXoEDMdRjOeKn8+vvkAwhviWXTHlG3R1QkbE5M/oywnZ83udJmi+lxjJ5 YhQ5IzomvJ16H0Bq+TLyVLO/VRksp1VR9HxCzItLNCS8PdpYYz5TC204ViycobYU65WMpzWe LFAGn8jSS25XIpqv0Y9k87dLbctKKA14Ifw2kq5OIVu2FuX+3i446JOa2vpCI9GcjCzi3oHV e00bzYiHMIl0FICrNJU0Kjho8pdo0m2uxkn6SYEpogAy9pnatUlO+erL4LqFUO7GXSdBRbw5 gNt25XTLdSFuZtMxkY3tq8MFss5QnjhehCVPEpE6y9ZjI4XB8ad1G4oBHVGK5LMsvg22PfMJ ISWFSHoF/B5+lHkCKWkFxZ0gZn33ju5n6/FOdEx4B8cMJt+cWwARAQABtClBbmRyZXcgQ29v cGVyIDxhbmRyZXcuY29vcGVyM0BjaXRyaXguY29tPokCOgQTAQgAJAIbAwULCQgHAwUVCgkI CwUWAgMBAAIeAQIXgAUCWKD95wIZAQAKCRBlw/kGpdefoHbdD/9AIoR3k6fKl+RFiFpyAhvO 59ttDFI7nIAnlYngev2XUR3acFElJATHSDO0ju+hqWqAb8kVijXLops0gOfqt3VPZq9cuHlh IMDquatGLzAadfFx2eQYIYT+FYuMoPZy/aTUazmJIDVxP7L383grjIkn+7tAv+qeDfE+txL4 SAm1UHNvmdfgL2/lcmL3xRh7sub3nJilM93RWX1Pe5LBSDXO45uzCGEdst6uSlzYR/MEr+5Z JQQ32JV64zwvf/aKaagSQSQMYNX9JFgfZ3TKWC1KJQbX5ssoX/5hNLqxMcZV3TN7kU8I3kjK mPec9+1nECOjjJSO/h4P0sBZyIUGfguwzhEeGf4sMCuSEM4xjCnwiBwftR17sr0spYcOpqET ZGcAmyYcNjy6CYadNCnfR40vhhWuCfNCBzWnUW0lFoo12wb0YnzoOLjvfD6OL3JjIUJNOmJy RCsJ5IA/Iz33RhSVRmROu+TztwuThClw63g7+hoyewv7BemKyuU6FTVhjjW+XUWmS/FzknSi dAG+insr0746cTPpSkGl3KAXeWDGJzve7/SBBfyznWCMGaf8E2P1oOdIZRxHgWj0zNr1+ooF /PzgLPiCI4OMUttTlEKChgbUTQ+5o0P080JojqfXwbPAyumbaYcQNiH1/xYbJdOFSiBv9rpt TQTBLzDKXok86LkCDQRS4TZ/ARAAkgqudHsp+hd82UVkvgnlqZjzz2vyrYfz7bkPtXaGb9H4 Rfo7mQsEQavEBdWWjbga6eMnDqtu+FC+qeTGYebToxEyp2lKDSoAsvt8w82tIlP/EbmRbDVn 7bhjBlfRcFjVYw8uVDPptT0TV47vpoCVkTwcyb6OltJrvg/QzV9f07DJswuda1JH3/qvYu0p vjPnYvCq4NsqY2XSdAJ02HrdYPFtNyPEntu1n1KK+gJrstjtw7KsZ4ygXYrsm/oCBiVW/OgU g/XIlGErkrxe4vQvJyVwg6YH653YTX5hLLUEL1NS4TCo47RP+wi6y+TnuAL36UtK/uFyEuPy wwrDVcC4cIFhYSfsO0BumEI65yu7a8aHbGfq2lW251UcoU48Z27ZUUZd2Dr6O/n8poQHbaTd 6bJJSjzGGHZVbRP9UQ3lkmkmc0+XCHmj5WhwNNYjgbbmML7y0fsJT5RgvefAIFfHBg7fTY/i kBEimoUsTEQz+N4hbKwo1hULfVxDJStE4sbPhjbsPCrlXf6W9CxSyQ0qmZ2bXsLQYRj2xqd1 bpA+1o1j2N4/au1R/uSiUFjewJdT/LX1EklKDcQwpk06Af/N7VZtSfEJeRV04unbsKVXWZAk uAJyDDKN99ziC0Wz5kcPyVD1HNf8bgaqGDzrv3TfYjwqayRFcMf7xJaL9xXedMcAEQEAAYkC HwQYAQgACQUCUuE2fwIbDAAKCRBlw/kGpdefoG4XEACD1Qf/er8EA7g23HMxYWd3FXHThrVQ HgiGdk5Yh632vjOm9L4sd/GCEACVQKjsu98e8o3ysitFlznEns5EAAXEbITrgKWXDDUWGYxd pnjj2u+GkVdsOAGk0kxczX6s+VRBhpbBI2PWnOsRJgU2n10PZ3mZD4Xu9kU2IXYmuW+e5KCA vTArRUdCrAtIa1k01sPipPPw6dfxx2e5asy21YOytzxuWFfJTGnVxZZSCyLUO83sh6OZhJkk b9rxL9wPmpN/t2IPaEKoAc0FTQZS36wAMOXkBh24PQ9gaLJvfPKpNzGD8XWR5HHF0NLIJhgg 4ZlEXQ2fVp3XrtocHqhu4UZR4koCijgB8sB7Tb0GCpwK+C4UePdFLfhKyRdSXuvY3AHJd4CP 4JzW0Bzq/WXY3XMOzUTYApGQpnUpdOmuQSfpV9MQO+/jo7r6yPbxT7CwRS5dcQPzUiuHLK9i nvjREdh84qycnx0/6dDroYhp0DFv4udxuAvt1h4wGwTPRQZerSm4xaYegEFusyhbZrI0U9tJ B8WrhBLXDiYlyJT6zOV2yZFuW47VrLsjYnHwn27hmxTC/7tvG3euCklmkn9Sl9IAKFu29RSo d5bD8kMSCYsTqtTfT6W4A3qHGvIDta3ptLYpIAOD2sY3GYq2nf3Bbzx81wZK14JdDDHUX2Rs 6+ahAA==
  • Cc: George Dunlap <George.Dunlap@xxxxxxxxxxxxx>, xen-devel <xen-devel@xxxxxxxxxxxxxxxxxxxx>, Wei Liu <wei.liu2@xxxxxxxxxx>, Roger Pau Monne <roger.pau@xxxxxxxxxx>
  • Delivery-date: Fri, 15 Mar 2019 16:54:12 +0000
  • List-id: Xen developer discussion <xen-devel.lists.xenproject.org>
  • Openpgp: preference=signencrypt

On 14/03/2019 17:15, Jan Beulich wrote:
>>>> On 14.03.19 at 17:38, <andrew.cooper3@xxxxxxxxxx> wrote:
>> On 19/12/2018 14:38, Jan Beulich wrote:
>>> @@ -8444,6 +8465,45 @@ x86_emulate(
>>>          generate_exception_if(ea.type != OP_MEM || !vex.l || vex.w, 
>>> EXC_UD);
>>>          goto simd_0f_avx2;
>>>  
>>> +    case X86EMUL_OPC_EVEX_66(0x0f38, 0x78): /* vpbroadcastb 
>>> xmm/m8,[xyz]mm{k} */
>>> +    case X86EMUL_OPC_EVEX_66(0x0f38, 0x79): /* vpbroadcastw 
>>> xmm/m16,[xyz]mm{k} */
>>> +        host_and_vcpu_must_have(avx512bw);
>>> +        generate_exception_if(evex.w || evex.brs, EXC_UD);
>>> +        op_bytes = elem_bytes = 1 << (b & 1);
>>> +        /* See the comment at the avx512_broadcast label. */
>>> +        op_mask |= !(b & 1 ? !(uint32_t)op_mask : !op_mask);
>>> +        goto avx512f_no_sae;
>>> +
>>> +    case X86EMUL_OPC_EVEX_66(0x0f38, 0x7a): /* vpbroadcastb r32,[xyz]mm{k} 
>>> */
>>> +    case X86EMUL_OPC_EVEX_66(0x0f38, 0x7b): /* vpbroadcastw r32,[xyz]mm{k} 
>>> */
>>> +        host_and_vcpu_must_have(avx512bw);
>>> +        generate_exception_if(evex.w, EXC_UD);
>>> +        /* fall through */
>>> +    case X86EMUL_OPC_EVEX_66(0x0f38, 0x7c): /* vpbroadcast{d,q} 
>>> reg,[xyz]mm{k} */
>>> +        generate_exception_if((ea.type != OP_REG || evex.brs ||
>>> +                               evex.reg != 0xf || !evex.RX),
>>> +                              EXC_UD);
>> generate_exception_if(ea.type != OP_REG || evex.brs ||
>>                       evex.reg != 0xf || !evex.RX, EXC_UD);
>>
>> ?
> Well, no - I don't really want the second argument on a continued
> line of the first one.

Why not?  We do this elsewhere, and it doesn't help readability to split
them.

>  Multiple full arguments on one line are fine
> with me. If you want me to drop just the inner parentheses, that
> would be fine (as mentioned in another context on this series, I've
> mainly added them because of what I understood your editor's
> behavior is when splitting lines like this, and I may have easily
> misunderstood).

Its only when your preferred indentation deviates from BSD style.  If
there are already brackets around, the chances are its fine, because the
align with brackets rule takes over.

The main issue is with return statements, where BSD style would require
the continuation of the statement to start under the second r of return.

A second more problematic issue is

exp ? one
    : two

because all indentation styles I'm aware of would move the second line
to left.  This one can't be fixed in any reasonable way by adding
brackets if there were previously none, because even if it was
bracketed, the else case would still move left.

>
>> Otherwise, Acked-by: Andrew Cooper <andrew.cooper3@xxxxxxxxxx>
> Let me know on what variant(s) of the above this holds.

I'd prefer not to split EXC_UD onto a line of its own, because I don't
think it helps the readability of the surrounding code, but I'm not
sufficiently bothered to make it a hard requirement.

~Andrew

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxxx
https://lists.xenproject.org/mailman/listinfo/xen-devel

 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.