Re: [Xen-devel] [xen-unstable-smoke test] 92827: regressions - FAIL

["Jan Beulich" <JBeulich@xxxxxxxx>]

>>> On 26.04.16 at 15:54, <Paul.Durrant@xxxxxxxxxx> wrote:
>>  -----Original Message-----
>> From: Wei Liu [mailto:wei.liu2@xxxxxxxxxx]
>> Sent: 26 April 2016 14:14
>> To: osstest service owner
>> Cc: xen-devel@xxxxxxxxxxxxxxxxxxx; Wei Liu; Jan Beulich; Paul Durrant;
>> Andrew Cooper
>> Subject: Re: [Xen-devel] [xen-unstable-smoke test] 92827: regressions - FAIL
>> 
>> CC Jan, Paul and Andrew
>> 
>> On Tue, Apr 26, 2016 at 01:01:32PM +0000, osstest service owner wrote:
>> > flight 92827 xen-unstable-smoke real [real]
>> > http://logs.test-lab.xenproject.org/osstest/logs/92827/ 
>> >
>> > Regressions :-(
>> >
>> > Tests which did not succeed and are blocking,
>> > including tests which could not be run:
>> >  test-amd64-amd64-xl-qemuu-debianhvm-i386 9 debian-hvm-install fail
>> REGR. vs. 92731
>> 
>> http://logs.test-lab.xenproject.org/osstest/logs/92827/test-amd64-amd64- 
>> xl-qemuu-debianhvm-i386/serial-fiano1.log
>> 
>> Apr 26 10:56:02.287445 (d1) HVM Loader
>> Apr 26 10:56:02.327467 (d1) Detected Xen v4.7-unstable
>> Apr 26 10:56:02.335490 (XEN) Assertion 'r->type == IOREQ_TYPE_COPY' failed
>> at vmsi.c:352
>> Apr 26 10:56:02.335531 (XEN) ----[ Xen-4.7-unstable  x86_64  debug=y  Not
>> tainted ]----
>> Apr 26 10:56:02.343504 (XEN) CPU:    3
>> Apr 26 10:56:02.343534 (XEN) RIP:    e008:[<ffff82d0801e2d3c>]
>> vmsi.c#msixtbl_range+0x6e/0xab
>> Apr 26 10:56:02.351504 (XEN) RFLAGS: 0000000000010297   CONTEXT:
>> hypervisor (d1v0)
>> Apr 26 10:56:02.359499 (XEN) rax: 000000000000000c   rbx: ffff82d08033c0a8
>> rcx: ffff83026bd0e608
>> Apr 26 10:56:02.367502 (XEN) rdx: 0000000000000000   rsi: 00000000003bfa0c
>> rdi: 0000000000000000
>> Apr 26 10:56:02.367543 (XEN) rbp: ffff830277eb7ae8   rsp: ffff830277eb7ac8
>> r8:  0000000000000001
>> Apr 26 10:56:02.375474 (XEN) r9:  00000000deadf00d   r10: ffff82d0802eebc0
>> r11: 0000000000000000
>> Apr 26 10:56:02.383492 (XEN) r12: 00000000003bfa0c   r13: ffff83007dce3000
>> r14: ffff830277eb0000
>> Apr 26 10:56:02.391501 (XEN) r15: ffff83007dce3000   cr0: 000000008005003b
>> cr4: 00000000001526e0
>> Apr 26 10:56:02.399556 (XEN) cr3: 000000026bc92000   cr2: 0000000000000000
>> Apr 26 10:56:02.399591 (XEN) ds: 0000   es: 0000   fs: 0000   gs: 0000   ss: 
> 0000
>> cs: e008
>> Apr 26 10:56:02.407619 (XEN) Xen code around <ffff82d0801e2d3c>
>> (vmsi.c#msixtbl_range+0x6e/0xab):
>> Apr 26 10:56:02.415619 (XEN)  00 bd 9f 09 00 00 01 74 <02> 0b 41 0f b6 85 9e 
> 09
>> 00 00 ba 00 00 00 00 a8
>> Apr 26 10:56:02.423608 (XEN) Xen stack trace from rsp=ffff830277eb7ac8:
>> Apr 26 10:56:02.423644 (XEN)    ffff830277eb7b58 ffff830259140170
>> 00000000003bfa0c ffff830277eb7b58
>> Apr 26 10:56:02.431612 (XEN)    ffff830277eb7b18 ffff82d0801d65ce
>> ffff830277eb7b18 ffff830259140170
>> Apr 26 10:56:02.439615 (XEN)    000000000000000b ffff83026bd0e000
>> ffff830277eb7b48 ffff82d0801d696e
>> Apr 26 10:56:02.447640 (XEN)    00000000003bfa0c 00000000003bfa0c
>> ffff820080000000 0000000000000008
>> Apr 26 10:56:02.455587 (XEN)    ffff830277eb7b78 ffff82d0801d6b2c
>> 00000000003bfa0c 0000000000000000
>> Apr 26 10:56:02.463604 (XEN)    0000000100000001 0100000000000000
>> ffff830277eb7c28 ffff82d0801cd3a1
>> Apr 26 10:56:02.463643 (XEN)    ffff830277eb7b98 ffff83007dce3000
>> ffff830277eb7bf4 ffff830277eb0000
>> Apr 26 10:56:02.471610 (XEN)    ffff830277eb7be4 ffff82d0801306d1
>> ffff830277eb7c08 0000000400000004
>> Apr 26 10:56:02.479606 (XEN)    ffff830277eb7c84 0000000000000a0c
>> 00000000000003bf 0000000100000008
>> Apr 26 10:56:02.487633 (XEN)    ffff82d08027ae11 ffff83026bd0eaa4
>> 0000000000000000 0000000000000008
>> Apr 26 10:56:02.495606 (XEN)    0000000000000009 ffff820080000000
>> ffff830277eb7f18 ffff820080000000
>> Apr 26 10:56:02.503633 (XEN)    ffff830277eb7c38 ffff82d0801cf050
>> ffff830277eb7c58 ffff82d0801cf48d
>> Apr 26 10:56:02.503674 (XEN)    0000000000000080 00000000003bfa0c
>> ffff830277eb7ce8 ffff82d08018a5f0
>> Apr 26 10:56:02.511615 (XEN)    0000000000000001 0000000000000000
>> ffff83026bd0e000 0000000077eb7e0a
>> Apr 26 10:56:02.519618 (XEN)    ffff830277eb7cb8 ffff82d0801cb478
>> ffff830277eb7d48 ffff830277eb7d68
>> Apr 26 10:56:02.527606 (XEN)    0000000000000000 ffff830259140190
>> ffff830277eb7cd8 ffff830277eb7f18
>> Apr 26 10:56:02.535607 (XEN)    ffff83007dce3000 0000000000000000
>> ffff830277eb7f18 ffff820080000000
>> Apr 26 10:56:02.543607 (XEN)    ffff830277eb7dc8 ffff82d08013d7db
>> ffff830277eb7d68 ffff830277eb7e0a
>> Apr 26 10:56:02.551600 (XEN)    ffff830277eb7d38 ffff830259140190
>> ffff830277eb7d68 ffff830277eb7d90
>> Apr 26 10:56:02.551638 (XEN)    ffff830277eb7d60 ffff830277eb7d68
>> ffff830277eb0000 ffff82d0801d69b9
>> Apr 26 10:56:02.559620 (XEN) Xen call trace:
>> Apr 26 10:56:02.559651 (XEN)    [<ffff82d0801e2d3c>]
>> vmsi.c#msixtbl_range+0x6e/0xab
>> Apr 26 10:56:02.567604 (XEN)    [<ffff82d0801d65ce>]
>> intercept.c#hvm_mmio_accept+0x6a/0xe4
>> Apr 26 10:56:02.575611 (XEN)    [<ffff82d0801d696e>]
>> hvm_find_io_handler+0x65/0x89
>> Apr 26 10:56:02.583602 (XEN)    [<ffff82d0801d6b2c>]
>> hvm_mmio_internal+0x37/0x61
> 
> Ah. Crap. I forgot about this path....

So did I. And my testing didn't catch it because I have a post-4.7
patch in place avoiding registration of the vMSI-X handler when
there are no passed through devices, and on the box where I do
use pass-through I normally run with debug=n.

>   I think the best way round this is to have vmsi register as an full io 
> interceptor so its accept method can use the passed in ioreq, which is faked 
> up to be a copy in this case. Either that or get rid of hvm_mmio_internal() 
> altogether.

Getting rid of it is not possible afaict. And converting to a full I/O
interceptor is a bad idea not just because of the backporting issue,
but also because we really don't _want_ to snoop accesses coming
through hvm_mmio_internal().

Jan

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel