test suite reviews and discussions
 help / color / mirror / Atom feed
From: Owen Hilyard <ohilyard@iol.unh.edu>
To: DongJunX <junx.dong@intel.com>
Cc: dts@dpdk.org, "Tu, Lijuan" <lijuan.tu@intel.com>, qingx.sun@intel.com
Subject: Re: [DTS][V2 1/2] framework/*: Add function to support ASan test
Date: Tue, 25 Jan 2022 16:10:30 -0500	[thread overview]
Message-ID: <CAHx6DYBjQO5-3zTMEYbr=_XveAWdjG4nN26oxRu1Y1Snhi1ScQ@mail.gmail.com> (raw)
In-Reply-To: <20220125083836.2354-2-junx.dong@intel.com>

[-- Attachment #1: Type: text/plain, Size: 1023 bytes --]

>
> sys.path.insert(0, DTS_PATH)
>

Please use python modules to handle this behavior. Directly modifying the
interpreter path causes issues with static analysis tools and IDEs.

Also, this seems to be designed to be entirely outside of the normal flow
for DTS. To me, it seems like you could do much the same thing by using
setting some variable (DTS_USE_ASAN) to true, and then in
the DPDKdut::build_install_dpdk_*_meson, add "-Db_sanitize=address" to the
arguments for meson. After all test suites are run, but before we output
results, add a check for that variable (DTS_USE_ASAN) and if it's true then
check the logs from all the test suites that were run for any ASAN failure
messages. If there is some reason why that isn't feasible, please let me
know. Currently, this seems like an alternative entry point to DTS that
will introduce an additional maintenance burden, and I think that making it
instead act as a few small behavioral changes in the existing DTS framework
would be easier to integrate with CI.

Owen

[-- Attachment #2: Type: text/html, Size: 1275 bytes --]

  reply	other threads:[~2022-01-25 21:11 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-01-25  8:38 [DTS][V2 0/2] " DongJunX
2022-01-25  8:38 ` [DTS][V2 1/2] framework/*: " DongJunX
2022-01-25 21:10   ` Owen Hilyard [this message]
2022-01-25  8:38 ` [DTS][V2 2/2] doc/*: Add ASan test user guide DongJunX

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to='CAHx6DYBjQO5-3zTMEYbr=_XveAWdjG4nN26oxRu1Y1Snhi1ScQ@mail.gmail.com' \
    --to=ohilyard@iol.unh.edu \
    --cc=dts@dpdk.org \
    --cc=junx.dong@intel.com \
    --cc=lijuan.tu@intel.com \
    --cc=qingx.sun@intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).