Please don't delay integration of this patch set further to port individual
tests. Individual tests can be ported to windows with significantly reduced
sensitivity to the churn once the conditional mechanisms are in place.
We spend enormous amounts of time continuously rebasing it every time one of
the existing tests is churned or new tests are added it's a losing game given
the rate of review the community provides for windows patches.
Thanks
-----Original Message-----
From: Stephen Hemminger <step...@networkplumber.org>
Sent: Wednesday, December 1, 2021 10:46 AM
To: Jie Zhou <j...@linux.microsoft.com>
Cc: dev@dpdk.org; Dmitry Kozlyuk <dmitry.kozl...@gmail.com>; Tyler Retzlaff
<roret...@microsoft.com>; navas...@linux.microsoft.com; Dmitry Malloy
<dmit...@microsoft.com>; Kadam, Pallavi <pallavi.ka...@intel.com>; talshn
<tal...@mellanox.com>; thomas <tho...@monjalon.net>; acon...@redhat.com
Subject: [EXTERNAL] Re: [PATCH v9 9/9] app/test: enable subset of unit tests on
Windows
On Wed, 1 Dec 2021 10:05:38 -0800
Jie Zhou <j...@linux.microsoft.com> wrote:
> diff --git a/app/test/test_dmadev.c b/app/test/test_dmadev.c index
> b206db27ae..c2ea9090cf 100644
> --- a/app/test/test_dmadev.c
> +++ b/app/test/test_dmadev.c
> @@ -2,6 +2,17 @@
> * Copyright(c) 2021 HiSilicon Limited
> * Copyright(c) 2021 Intel Corporation
> */
> +#include "test.h"
> +
> +#ifdef RTE_EXEC_ENV_WINDOWS
> +static int
> +test_dma(void)
> +{
> + printf("dma not supported on Windows, skipping test\n");
> + return TEST_SKIPPED;
> +}
> +
> +#else
Rather than littering code with ifdefs is it not possible to just not build
these tests on Windows?
