[11u] RFR: 8226406: JVM fails to detect mismatched or corrupt CDS archive

[Lindenmaier, Goetz]

Hi Jiangli,

thanks for your review! I will add a comment. 
I thought better leave this in, in case the other functionality
will be downported one can see this is needed, too.

Best regards,
  Goetz

> -----Original Message-----
> From: Jiangli Zhou <jianglizhou at google.com>
> Sent: Mittwoch, 29. Januar 2020 00:20
> To: Lindenmaier, Goetz <goetz.lindenmaier at sap.com>
> Cc: jdk-updates-dev at openjdk.java.net; hotspot-runtime-dev at openjdk.java.net
> Subject: Re: [11u] RFR: 8226406: JVM fails to detect mismatched or corrupt
> CDS archive
> 
> Hi Goetz,
> 
> The backport and modifications in filemap.cpp look good. Would it be
> good to also add a comment noting the code below is backported from
> JDK 13 and not compatible/needed for 11?
> 
>  527   //if (_header->_header_size != sz) {
>  528   //  log_info(cds)("_header_size expected: " SIZE_FORMAT, sz);
>  529   //  log_info(cds)("               actual: " SIZE_FORMAT,
> _header->_header_size);
>  530   //  FileMapInfo::fail_continue("The shared archive file has an
> incorrect header size.");
>  531   //  return false;
>  532   // }
> 
> I'll also run some tests on my side with your patch applied.
> 
> Best regards,
> 
> Jiangli
> 
> On Tue, Jan 28, 2020 at 7:07 AM Lindenmaier, Goetz
> <goetz.lindenmaier at sap.com> wrote:
> >
> > Hi
> >
> > I would like to downport gthis change for parity with 11.0.7-oracle.
> >
> > I had to do several adaptions of the change to get the code into 11u.
> >
> > In filemap.cpp jdk13 distinguishes CDS_ARCHIVE_MAGIC and
> CDS_DYNAMIC_ARCHIVE_MAGIC. Also, 13 checks field _header_size which is
> not in 11.
> >
> > I simplified this to check only for CDS_ARCHIVE_MAGIC which is known in 11.
> I removed the checks for _header_size.
> >
> > The test too varies a lot between 11 and 13.
> >
> > Ran it through our testing, and ran the test manually. Both passed.
> >
> > http://cr.openjdk.java.net/~goetz/wr20/8226406-corrupt_CDS_archive-
> jdk11/01/
> >
> > Best regards,
> >   Goetz