netdev - Re: [PATCH v5 3/3] vsock/test: verify socket options after setting them

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <jjplqknxiasru7oukfrz66eryfhtviukwchxkh4yq6woj3m7qw@augnv6adwrgc>
Date: Tue, 12 Nov 2024 19:10:21 +0100
From: Stefano Garzarella <sgarzare@...hat.com>
To: Konstantin Shkolnyy <kshk@...ux.ibm.com>
Cc: virtualization@...ts.linux.dev, netdev@...r.kernel.org, 
	linux-kernel@...r.kernel.org, mjrosato@...ux.ibm.com
Subject: Re: [PATCH v5 3/3] vsock/test: verify socket options after setting
 them

On Tue, Nov 12, 2024 at 09:18:48AM -0600, Konstantin Shkolnyy wrote:
>On 11/12/2024 02:58, Stefano Garzarella wrote:
>>On Thu, Nov 07, 2024 at 07:17:26PM -0600, Konstantin Shkolnyy wrote:
>>>Replace setsockopt() calls with calls to functions that follow
>>>setsockopt() with getsockopt() and check that the returned value and its
>>>size are the same as have been set.
>>>
>>>Signed-off-by: Konstantin Shkolnyy <kshk@...ux.ibm.com>
>>>---
>>>tools/testing/vsock/Makefile              |   8 +-
>>>tools/testing/vsock/control.c             |   8 +-
>>>tools/testing/vsock/msg_zerocopy_common.c |   8 +-
>>>tools/testing/vsock/util_socket.c         | 149 ++++++++++++++++++++++
>>>tools/testing/vsock/util_socket.h         |  19 +++
>>>tools/testing/vsock/vsock_perf.c          |  24 ++--
>>>tools/testing/vsock/vsock_test.c          |  40 +++---
>>>7 files changed, 208 insertions(+), 48 deletions(-)
>>>create mode 100644 tools/testing/vsock/util_socket.c
>>>create mode 100644 tools/testing/vsock/util_socket.h
>>>
>>>diff --git a/tools/testing/vsock/Makefile b/tools/testing/vsock/Makefile
>>>index 6e0b4e95e230..1ec0b3a67aa4 100644
>>>--- a/tools/testing/vsock/Makefile
>>>+++ b/tools/testing/vsock/Makefile
>>>@@ -1,12 +1,12 @@
>>># SPDX-License-Identifier: GPL-2.0-only
>>>all: test vsock_perf
>>>test: vsock_test vsock_diag_test vsock_uring_test
>>>-vsock_test: vsock_test.o vsock_test_zerocopy.o timeout.o 
>>>control.o util.o msg_zerocopy_common.o
>>>-vsock_diag_test: vsock_diag_test.o timeout.o control.o util.o
>>>-vsock_perf: vsock_perf.o msg_zerocopy_common.o
>>>+vsock_test: vsock_test.o vsock_test_zerocopy.o timeout.o 
>>>control.o util.o msg_zerocopy_common.o util_socket.o
>>>+vsock_diag_test: vsock_diag_test.o timeout.o control.o util.o 
>>>util_socket.o
>>>+vsock_perf: vsock_perf.o msg_zerocopy_common.o util_socket.o
>>
>>I would add the new functions to check setsockopt in util.c
>>
>>vsock_perf is more of a tool to measure performance than a test, so
>>we can avoid calling these checks there, tests should cover all
>>cases regardless of vsock_perf.
>
>The problem is that vsock_perf calls enable_so_zerocopy() which has to
>call the new setsockopt_int_check() because it's also called by 
>vsock_test. Do you prefer to give vsock_perf its own version of
>enable_so_zerocopy() which doesn't call setsockopt_int_check()?
>

Yeah, maybe we can move the old enable_so_zerocopy() in vsock_perf.c
and implement another enable_so_zerocopy() in util.c for the tests.

Thanks,
Stefano