[ 
https://issues.apache.org/jira/browse/AVRO-4248?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18075569#comment-18075569
 ] 

Andrew Pilloud commented on AVRO-4248:
--------------------------------------

I believe this is purely a sanitizer issue on x86, it might be a crash on other 
architectures but I don't have any examples.

> ASAN misaligned address in BufferDetail.hh
> ------------------------------------------
>
>                 Key: AVRO-4248
>                 URL: https://issues.apache.org/jira/browse/AVRO-4248
>             Project: Apache Avro
>          Issue Type: Bug
>          Components: c++
>    Affects Versions: 1.8.0
>            Reporter: Andrew Pilloud
>            Priority: Trivial
>              Labels: pull-request-available
>          Time Spent: 10m
>  Remaining Estimate: 0h
>
> We're on a really old version of Avro, but this appears to still exist at 
> head.
> avro/api/buffer/detail/BufferDetail.hh:355:13: runtime error: store to 
> misaligned address 0x7d5a4dde650f for type 'long', which requires 8 byte 
> alignment
> 0x7d5a4dde650f: note: pointer points here
>  72 22 00 04 ab  ab ab ab ab ab ab ab ab  ab ab ab ab ab ab ab ab  ab ab ab 
> ab ab ab ab ab  ab ab ab
>              ^ 
>     #0 0x7f4b2c7b93bf in void avro::detail::BufferImpl::writeTo<long>(long, 
> boost::integral_constant<bool, true> const&) 
> avro/api/buffer/detail/BufferDetail.hh:355:78
>     #1 0x7f4b2c7b40d1 in writeTo<long> avro/api/buffer/Buffer.hh:141:17
>     #2 0x7f4b2c7b40d1 in writeValue avro/api/Writer.hh:90:17
>     #3 0x7f4b2c7b40d1 in writeDouble avro/api/Serializer.hh:71:17
>     #4 0x7f4b2c7b40d1 in void 
> TestSchema::writeEncoding<avro::Serializer<avro::WriterImpl<avro::NullValidator>>>(avro::Serializer<avro::WriterImpl<avro::NullValidator>>&,
>  int) third_party/avro/test/unittest.cc:245:11
>     #5 0x7f4b2c79e8d9 in TestSchema::printEncoding() 
> avro/test/unittest.cc:278:9
>     #6 0x7f4b2c73637c in TestSchema::test() avro/test/unittest.cc:545:9



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

Reply via email to