All of lore.kernel.org
 help / color / mirror / Atom feed
From: Ian Jackson <iwj@xenproject.org>
To: Andrew Cooper <andrew.cooper3@citrix.com>
Cc: Xen-devel <xen-devel@lists.xenproject.org>,
	Ian Jackson <iwj@xenproject.org>, Wei Liu <wl@xen.org>,
	Anthony PERARD <anthony.perard@citrix.com>
Subject: Re: [PATCH 04/10] tools/libxl: Fix uninitialised variable in libxl__domain_get_device_model_uid()
Date: Tue, 16 Feb 2021 16:07:01 +0000	[thread overview]
Message-ID: <24619.60837.362896.565993@mariner.uk.xensource.com> (raw)
In-Reply-To: <20210212153953.4582-5-andrew.cooper3@citrix.com>

Andrew Cooper writes ("[PATCH 04/10] tools/libxl: Fix uninitialised variable in libxl__domain_get_device_model_uid()"):
> The logic is sufficiently complicated I can't figure out if the complain is
> legitimate or not.  There is exactly one path wanting kill_by_uid set to true,
> so default it to false and drop the existing workaround for this problem at
> other optimisation levels.

The place where it's used is here:

    if (!rc && user) {
            state->dm_runas = user;
                    if (kill_by_uid)
                                state->dm_kill_uid = GCSPRINTF("%ld",...
        
This is gated by !rc.  So for this to be used uninitialised, we'd have
to get here with rc==0 but uninitialised kill_by_uid.

The label `out` is preceded by a nonzero assignment to rc.

All the `goto out` are preceded by either (i) nonzero assignment to
rc, or (ii) assignment to kill_by_uid and setting rc=0.

So the compiler is wrong.

If only we had sum types.

In the absence of sum types I suggest the following restructuring:
Change all the `rc = ERROR...; goto out;` to `goto err` and make `goto
out` be the success path only.

Ian.


  reply	other threads:[~2021-02-16 16:07 UTC|newest]

Thread overview: 47+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-02-12 15:39 [PATCH for-4.15 00/10] tools: Support to use abi-dumper on libraries Andrew Cooper
2021-02-12 15:39 ` [PATCH 01/10] tools/xl: Fix exit code for `xl vkbattach` Andrew Cooper
2021-02-16 15:57   ` Ian Jackson
2021-02-16 16:25   ` Ian Jackson
2021-02-12 15:39 ` [PATCH 02/10] tools/libxg: Fix uninitialised variable in write_x86_cpu_policy_records() Andrew Cooper
2021-02-12 15:54   ` Jan Beulich
2021-02-16 15:57   ` Ian Jackson
2021-02-12 15:39 ` [PATCH 03/10] tools/libxg: Fix uninitialised variable in meminit() Andrew Cooper
2021-02-12 15:55   ` Julien Grall
2021-02-12 19:35     ` Andrew Cooper
2021-02-12 20:01   ` [PATCH v1.1 03/10] tools/libxg: Drop stale p2m logic from ARM's meminit() Andrew Cooper
2021-02-16 16:00     ` Ian Jackson
2021-02-16 16:27     ` Julien Grall
2021-02-12 15:39 ` [PATCH 04/10] tools/libxl: Fix uninitialised variable in libxl__domain_get_device_model_uid() Andrew Cooper
2021-02-16 16:07   ` Ian Jackson [this message]
2021-02-16 17:43   ` [PATCH v2 " Andrew Cooper
2021-02-16 17:57     ` Ian Jackson
2021-02-16 18:05       ` Ian Jackson
2021-02-12 15:39 ` [PATCH 05/10] tools/libxl: Fix uninitialised variable in libxl__write_stub_dmargs() Andrew Cooper
2021-02-16 16:07   ` Ian Jackson
2021-02-16 16:26   ` Ian Jackson
2021-02-12 15:39 ` [PATCH 06/10] stubdom/xenstored: Fix uninitialised variables in lu_read_state() Andrew Cooper
2021-02-12 16:08   ` Jürgen Groß
2021-02-12 17:01     ` Andrew Cooper
2021-02-13  9:36       ` Jürgen Groß
2021-02-15 14:12         ` Andrew Cooper
2021-02-15 15:36   ` [PATCH v1.1 " Andrew Cooper
2021-02-16 16:10   ` [PATCH " Ian Jackson
2021-02-12 15:39 ` [PATCH 07/10] tools: Use -Og for debug builds when available Andrew Cooper
2021-02-12 16:04   ` Jan Beulich
2021-02-12 16:09     ` Andrew Cooper
2021-02-12 16:14       ` Jan Beulich
2021-02-16 16:26     ` Ian Jackson
2021-02-16 16:11   ` Ian Jackson
2021-02-12 15:39 ` [PATCH 08/10] tools: Check for abi-dumper in ./configure Andrew Cooper
2021-02-16 16:11   ` Ian Jackson
2021-02-16 16:27   ` Ian Jackson
2021-02-12 15:39 ` [PATCH 09/10] tools/libs: Add rule to generate headers.lst Andrew Cooper
2021-02-16 16:13   ` Ian Jackson
2021-02-16 16:48   ` [PATCH v2 " Andrew Cooper
2021-02-16 17:32     ` Ian Jackson
2021-02-12 15:39 ` [PATCH 10/10] tools/libs: Write out an ABI analysis when abi-dumper is available Andrew Cooper
2021-02-12 16:12   ` Jan Beulich
2021-02-12 17:03     ` Andrew Cooper
2021-02-12 18:01   ` [PATCH v1.1 " Andrew Cooper
2021-02-16 16:15   ` [PATCH " Ian Jackson
2021-02-16 16:30   ` Ian Jackson

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=24619.60837.362896.565993@mariner.uk.xensource.com \
    --to=iwj@xenproject.org \
    --cc=andrew.cooper3@citrix.com \
    --cc=anthony.perard@citrix.com \
    --cc=wl@xen.org \
    --cc=xen-devel@lists.xenproject.org \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.