Re: [dpdk-dev] [RFC v2] porting AddressSanitizer feature to DPDK

["Lin, Xueqin" <xueqin.lin@intel.com>]

> -----Original Message-----
> From: Jerin Jacob <jerinjacobk@gmail.com>
> Sent: Wednesday, June 16, 2021 7:35 PM
> To: Lin, Xueqin <xueqin.lin@intel.com>
> Cc: Peng, ZhihongX <zhihongx.peng@intel.com>; Burakov, Anatoly
> <anatoly.burakov@intel.com>; Ananyev, Konstantin
> <konstantin.ananyev@intel.com>; Stephen Hemminger
> <stephen@networkplumber.org>; dpdk-dev <dev@dpdk.org>
> Subject: Re: [dpdk-dev] [RFC v2] porting AddressSanitizer feature to DPDK
> 
> On Wed, Jun 16, 2021 at 2:43 PM Lin, Xueqin <xueqin.lin@intel.com> wrote:
> >
> > > -----Original Message-----
> > > From: Jerin Jacob <jerinjacobk@gmail.com>
> > > Sent: Tuesday, June 15, 2021 4:40 PM
> > > To: Peng, ZhihongX <zhihongx.peng@intel.com>
> > > Cc: Burakov, Anatoly <anatoly.burakov@intel.com>; Ananyev,
> > > Konstantin <konstantin.ananyev@intel.com>; Stephen Hemminger
> > > <stephen@networkplumber.org>; dpdk-dev <dev@dpdk.org>; Lin,
> Xueqin
> > > <xueqin.lin@intel.com>
> > > Subject: Re: [dpdk-dev] [RFC v2] porting AddressSanitizer feature to
> > > DPDK
> > >
> > > On Tue, Jun 15, 2021 at 1:46 PM <zhihongx.peng@intel.com> wrote:
> > > >
> > > > From: Zhihong Peng <zhihongx.peng@intel.com>
> > > >
> > > > AddressSanitizer (ASan) is a google memory error detect standard tool.
> > > > It could help to detect use-after-free and
> > > > {heap,stack,global}-buffer overflow bugs in C/C++ programs, print
> > > > detailed error information when error happens, large improve debug
> efficiency.
> > > >
> > > > By referring to its implementation algorithm
> > > > (https://github.com/google/sanitizers/wiki/AddressSanitizerAlgorit
> > > > hm), ported heap-buffer-overflow and use-after-freefunctions to
> > > > dpdk.
> > > >
> > > > Here is an example of heap-buffer-overflow bug:
> > > >         ......
> > > >         char *p = rte_zmalloc(NULL, 7, 0);
> > > >         p[7] = 'a';
> > > >         ......
> > > >
> > > > Here is an example of use-after-free bug:
> > > >         ......
> > > >         char *p = rte_zmalloc(NULL, 7, 0);
> > > >         rte_free(p);
> > > >         *p = 'a';
> > > >         ......
> > > >
> > > > If you want to use this feature,
> > > > you need to use the following compilation options:
> > > > -Db_lundef=false -Db_sanitize=address
> > >
> > > # Thanks for this patch. It is a useful item.
> > >
> > > # Subject could be changed
> > > from:
> > > porting AddressSanitizer feature to DPDK to
> > > eal: support for  AddressSanitizer
> > > or so
> >
> > Thanks for your positive feedback and review.
> > Good point, we will update the title in next version.
> >
> > >
> > > # Could you add a section in the documentation for Sanitizers to
> > > document the build time option and other points that users need to know.
> >
> > Make sense to add build option and key points to document, will add
> > this part in doc
> folder.
> >
> > > We can add other sanitizers such as UBSan etc in the future here
> > WIP to research other sanitizer tool.
> 
> UBsan is a good candate.
> Some old DPDK patch for the same:
> http://patches.dpdk.org/project/dpdk/patch/1573832013-18946-1-git-send-
> email-hkalra@marvell.com/
> 

Thanks for share, we will research the tool next. 

> > >
> > > # Add a UT test case to make sure it is working in app/test or so.
> >
> > This tool could help to detect memory issue, need to change bad code to
> check if working.
> 
> It is better to have a  UT to test things are working. You could add it in
> app/test/test_address_sanity.c. Tests can be such that
> - Skip if not complied with Sanity enabled
> - Pass if the code detects the known bad code. You can have test cases with
> pubic rte_ API that internally exercise the verify your implementation related
> to new asan_* APIs.

Thanks Jacob for your suggestion, we will add unit test for Asan tool later. 

> 
> 
> > Suggest listing demo code and tool capture information for user to try if
> tool works, also add this part into doc.
> >
> > >
> > > # Also, Please update the release note for this feature.
> > Sure, we can update the release note if code merge.
> 
> Probably you can send v1 version next i.e change the RFC status to get
> merged.

Sure, we will send v1 patch if no obvious objection for that, hope patch could receive some ACKs and could success to merge, thanks.  

> 
> 
> >