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

Re: [Xen-devel] [PATCH] page-alloc: detect double free earlier


  • To: Jan Beulich <JBeulich@xxxxxxxx>
  • From: Andrew Cooper <andrew.cooper3@xxxxxxxxxx>
  • Date: Thu, 9 May 2019 14:34:09 +0100
  • Authentication-results: esa1.hc3370-68.iphmx.com; dkim=none (message not signed) header.i=none; spf=None smtp.pra=andrew.cooper3@xxxxxxxxxx; spf=SoftFail smtp.mailfrom=Andrew.Cooper3@xxxxxxxxxx; spf=None smtp.helo=postmaster@xxxxxxxxxxxxxxxxxxxxxxxxxx
  • 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: Stefano Stabellini <sstabellini@xxxxxxxxxx>, Wei Liu <wei.liu2@xxxxxxxxxx>, Konrad Rzeszutek Wilk <konrad.wilk@xxxxxxxxxx>, George Dunlap <George.Dunlap@xxxxxxxxxxxxx>, Tim Deegan <tim@xxxxxxx>, Ian Jackson <Ian.Jackson@xxxxxxxxxxxxx>, Julien Grall <julien.grall@xxxxxxx>, xen-devel <xen-devel@xxxxxxxxxxxxxxxxxxxx>
  • Delivery-date: Thu, 09 May 2019 14:04:29 +0000
  • Ironport-sdr: JMpQM32534b8MMq6FfUVXm1k8Q5GonKR3ugQZH1JBQB93lSG5uk94Uf5yE9YrtGUqqbsI4ZIM2 nJnIIwm8NLditMz4zpTpCvBsiiUDSpRtVAIbQSsrLSfW6ddcn4rSg0gSQsX6ew3KwwfotWtszn OdIDjCex+usIzJqikNABwjbBQXOTQ/rUOGcORKsh42BRzPMirZ7i8QEDMMWXrWFD2EZHIFKl5h J2fDO8qw4FNC+++R/lM4VKtVnYDo4lrVloo5Vt30U9qfWhnbyV2BhMSgc00XoH0OMHwidHJVm8 Eek=
  • List-id: Xen developer discussion <xen-devel.lists.xenproject.org>
  • Openpgp: preference=signencrypt

On 09/05/2019 14:25, Jan Beulich wrote:
>>>> On 09.05.19 at 14:50, <andrew.cooper3@xxxxxxxxxx> wrote:
>> On 09/05/2019 13:35, Jan Beulich wrote:
>>> Right now this goes unnoticed until some subsequent page allocator
>>> operation stumbles across the thus corrupted list. We can do better:
>>> Only PGC_state_inuse and PGC_state_offlining pages can legitimately be
>>> passed to free_heap_pages().
>>>
>>> Take the opportunity and also restrict the PGC_broken check to the
>>> PGC_state_offlining case, as only pages of that type or
>>> PGC_state_offlined may have this flag set on them. Similarly, since
>>> PGC_state_offlined is not a valid input state, the setting of "tainted"
>>> can be restricted to just this case.
>>>
>>> Signed-off-by: Jan Beulich <jbeulich@xxxxxxxx>
>> Acked-by: Andrew Cooper <andrew.cooper3@xxxxxxxxxx>, with a suggestion.
> Thanks.
>
>>> --- a/xen/common/page_alloc.c
>>> +++ b/xen/common/page_alloc.c
>>> @@ -1409,13 +1409,22 @@ static void free_heap_pages(
>>>           *     in its pseudophysical address space).
>>>           * In all the above cases there can be no guest mappings of this 
>>> page.
>>>           */
>>> -        ASSERT(!page_state_is(&pg[i], offlined));
>>> -        pg[i].count_info =
>>> -            ((pg[i].count_info & PGC_broken) |
>>> -             (page_state_is(&pg[i], offlining)
>>> -              ? PGC_state_offlined : PGC_state_free));
>>> -        if ( page_state_is(&pg[i], offlined) )
>>> +        switch ( pg[i].count_info & PGC_state )
>>> +        {
>>> +        case PGC_state_inuse:
>>> +            BUG_ON(pg[i].count_info & PGC_broken);
>>> +            pg[i].count_info = PGC_state_free;
>>> +            break;
>>> +
>>> +        case PGC_state_offlining:
>>> +            pg[i].count_info = (pg[i].count_info & PGC_broken) |
>>> +                               PGC_state_offlined;
>>>              tainted = 1;
>>> +            break;
>>> +
>>> +        default:
>> Given that this is a fully fatal condition, it would be helpful to at
>> least print the state we found here.  For cases other than
>> PGC_state_free, it would probably be a very useful piece of information
>> for diagnosing what went wrong.
> Funny you should say this - I have the debugging patch below on top
> in my tree. I could easily submit this as a standalone follow-on patch.

TBH, I think it would be fine folded in, although with...

>
> Jan
>
> --- unstable.orig/xen/common/page_alloc.c
> +++ unstable/xen/common/page_alloc.c
> @@ -1014,7 +1014,14 @@ static struct page_info *alloc_heap_page
>      for ( i = 0; i < (1 << order); i++ )
>      {
>          /* Reference count must continuously be zero for free pages. */
> -        BUG_ON((pg[i].count_info & ~PGC_need_scrub) != PGC_state_free);
> +        if ( (pg[i].count_info & ~PGC_need_scrub) != PGC_state_free )
> +        {
> +            printk(XENLOG_ERR "pg[%x] m=%lx c=%lx o=%x v=%lx t=%x\n",

"pg[%u] mfn %"PRImfn" c=%#lx o=%u v=%#lx t=%#x\n"

so we don't end up printing numbers which are ambiguous between hex/dec.

With at least the ambiguity removed, my ack stands.

~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®.