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

Re: [Xen-devel] [PATCH 1/2] x86/svm: Improve diagnostics when __get_instruction_length_from_list() fails


  • To: Jan Beulich <JBeulich@xxxxxxxx>
  • From: Andrew Cooper <andrew.cooper3@xxxxxxxxxx>
  • Date: Mon, 3 Dec 2018 11:45:21 +0000
  • Autocrypt: addr=andrew.cooper3@xxxxxxxxxx; prefer-encrypt=mutual; keydata= xsFNBFLhNn8BEADVhE+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+cWwARAQABzSlBbmRyZXcgQ29v cGVyIDxhbmRyZXcuY29vcGVyM0BjaXRyaXguY29tPsLBegQTAQgAJAIbAwULCQgHAwUVCgkI 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 TQTBLzDKXok86M7BTQRS4TZ/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 uAJyDDKN99ziC0Wz5kcPyVD1HNf8bgaqGDzrv3TfYjwqayRFcMf7xJaL9xXedMcAEQEAAcLB XwQYAQgACQUCUuE2fwIbDAAKCRBlw/kGpdefoG4XEACD1Qf/er8EA7g23HMxYWd3FXHThrVQ HgiGdk5Yh632vjOm9L4sd/GCEACVQKjsu98e8o3ysitFlznEns5EAAXEbITrgKWXDDUWGYxd pnjj2u+GkVdsOAGk0kxczX6s+VRBhpbBI2PWnOsRJgU2n10PZ3mZD4Xu9kU2IXYmuW+e5KCA vTArRUdCrAtIa1k01sPipPPw6dfxx2e5asy21YOytzxuWFfJTGnVxZZSCyLUO83sh6OZhJkk b9rxL9wPmpN/t2IPaEKoAc0FTQZS36wAMOXkBh24PQ9gaLJvfPKpNzGD8XWR5HHF0NLIJhgg 4ZlEXQ2fVp3XrtocHqhu4UZR4koCijgB8sB7Tb0GCpwK+C4UePdFLfhKyRdSXuvY3AHJd4CP 4JzW0Bzq/WXY3XMOzUTYApGQpnUpdOmuQSfpV9MQO+/jo7r6yPbxT7CwRS5dcQPzUiuHLK9i nvjREdh84qycnx0/6dDroYhp0DFv4udxuAvt1h4wGwTPRQZerSm4xaYegEFusyhbZrI0U9tJ B8WrhBLXDiYlyJT6zOV2yZFuW47VrLsjYnHwn27hmxTC/7tvG3euCklmkn9Sl9IAKFu29RSo d5bD8kMSCYsTqtTfT6W4A3qHGvIDta3ptLYpIAOD2sY3GYq2nf3Bbzx81wZK14JdDDHUX2Rs 6+ahAA==
  • Cc: Wei Liu <wei.liu2@xxxxxxxxxx>, Xen-devel <xen-devel@xxxxxxxxxxxxx>, Paul Durrant <paul.durrant@xxxxxxxxxx>, Suravee Suthikulpanit <suravee.suthikulpanit@xxxxxxx>, Boris Ostrovsky <boris.ostrovsky@xxxxxxxxxx>, Brian Woods <brian.woods@xxxxxxx>, Roger Pau Monne <roger.pau@xxxxxxxxxx>
  • Delivery-date: Mon, 03 Dec 2018 11:45:31 +0000
  • List-id: Xen developer discussion <xen-devel.lists.xenproject.org>
  • Openpgp: preference=signencrypt

On 03/12/2018 10:32, Jan Beulich wrote:
>>>> On 30.11.18 at 18:07, <andrew.cooper3@xxxxxxxxxx> wrote:
>> Also, I'm not entirely convinced that making modrm an annonymous union is
>> going to work with older CentOS compilers,
> It certainly won't.
>
>> and therefore am not sure whether
>> that part of the change is worth it.  The instruction in question can be
>> obtained from the printed INSN_ constant alone.
>> ---
>>  xen/arch/x86/hvm/svm/emulate.c | 26 +++++++++++++++++++-------
>>  1 file changed, 19 insertions(+), 7 deletions(-)
>>
>> diff --git a/xen/arch/x86/hvm/svm/emulate.c b/xen/arch/x86/hvm/svm/emulate.c
>> index 3d04af0..71a1b6e 100644
>> --- a/xen/arch/x86/hvm/svm/emulate.c
>> +++ b/xen/arch/x86/hvm/svm/emulate.c
>> @@ -56,11 +56,14 @@ static unsigned long svm_nextrip_insn_length(struct vcpu 
>> *v)
>>  
>>  static const struct {
>>      unsigned int opcode;
>> -    struct {
>> -        unsigned int rm:3;
>> -        unsigned int reg:3;
>> -        unsigned int mod:2;
>> -#define MODRM(mod, reg, rm) { rm, reg, mod }
>> +    union {
>> +        struct {
>> +            unsigned int rm:3;
>> +            unsigned int reg:3;
>> +            unsigned int mod:2;
>> +        };
>> +        unsigned int raw;
> Why unsigned int instead of uint8_t?

Because to being with, this was a diagnostic patch and copied the type
above without thinking too much.

>
>> @@ -152,8 +155,17 @@ int __get_instruction_length_from_list(struct vcpu *v,
>>      }
>>  
>>      gdprintk(XENLOG_WARNING,
>> -             "%s: Mismatch between expected and actual instruction: "
>> -             "eip = %lx\n",  __func__, (unsigned long)vmcb->rip);
>> +             "%s: Mismatch between expected and actual instruction:\n",
>> +             __func__);
>> +    gdprintk(XENLOG_WARNING,
>> +             "  list[0] val %d, { opc %#x, modrm %#x }, list entries: %u\n",
>> +             list[0], opc_tab[list[0]].opcode, opc_tab[list[0]].modrm.raw,
>> +             list_count);
>> +    gdprintk(XENLOG_WARNING, "  rip 0x%lx, nextrip 0x%lx, len %lu\n",
>> +             vmcb->rip, vmcb->nextrip, vmcb->nextrip - vmcb->rip);
>> +    hvm_dump_emulation_state(XENLOG_G_WARNING, "Insn_len",
>> +                             &ctxt, X86EMUL_UNHANDLEABLE);
>> +
>>      hvm_inject_hw_exception(TRAP_gp_fault, 0);
>>      return 0;
>>  }
> The gdprintk()s all expanding to nothing in release builds I'm
> not fully convinced the added verbosity is worth it. In debug
> builds adding some debugging code like this shouldn't be a
> big hurdle.

You and I know what diagnostics to put here, but I think Pauls reaction
to finding that message demonstrates that most others don't.  Nor should
they - the peculiarities of first-gen AMD hardware needn't be mandatory
knowledge for most contributors.

For these diagnostics, they are only reachable in debug builds, and this
codepath is only even reachable in release builds on first-gen hardware.

However, the difference between what is currently present and this is
enough information to actually diagnose the problem.

~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®.