Re: [PATCH v2 2/4] virtio_test: move magic number in code as defined constant

From: "Michael S. Tsirkin" <mst@redhat.com>
To: Guo Zhi <qtxuning1999@sjtu.edu.cn>
Cc: jasowang@redhat.com, eperezma@redhat.com,
	virtualization@lists.linux-foundation.org,
	linux-kernel@vger.kernel.org, sgarzare@redhat.com
Subject: Re: [PATCH v2 2/4] virtio_test: move magic number in code as defined constant
Date: Thu, 7 Jul 2022 01:09:55 -0400	[thread overview]
Message-ID: <20220707010036-mutt-send-email-mst@kernel.org> (raw)
In-Reply-To: <20220707024409.1869-3-qtxuning1999@sjtu.edu.cn>

On Thu, Jul 07, 2022 at 10:44:07AM +0800, Guo Zhi wrote:
> We should avoid using magic numbers directly.

Not necessarily. For repeated values - I guess.
But this kind of thing:

	#define BUF_SIZE 1024
	int buf_size = BUF_SIZE;

brings no benefit IMHO.

And this has cost - values are now removed from code.

> Signed-off-by: Guo Zhi <qtxuning1999@sjtu.edu.cn>
> ---
>  tools/virtio/virtio_test.c | 18 +++++++++++-------
>  1 file changed, 11 insertions(+), 7 deletions(-)
> 
> diff --git a/tools/virtio/virtio_test.c b/tools/virtio/virtio_test.c
> index 95f78b311..1ecd64271 100644
> --- a/tools/virtio/virtio_test.c
> +++ b/tools/virtio/virtio_test.c
> @@ -20,7 +20,10 @@
>  #include "../../drivers/vhost/test.h"
>  
>  #define RANDOM_BATCH -1
> -
> +#define ALIGN 4096
> +#define RINGSIZE   256
> +#define TEST_BUF_NUM 0x100000
> +#define BUF_SIZE   1024
>  /* Unused */
>  void *__kmalloc_fake, *__kfree_ignore_start, *__kfree_ignore_end;
>  
> @@ -100,8 +103,8 @@ static void vq_reset(struct vq_info *info, int num, struct virtio_device *vdev)
>  	if (info->vq)
>  		vring_del_virtqueue(info->vq);
>  
> -	memset(info->ring, 0, vring_size(num, 4096));
> -	vring_init(&info->vring, num, info->ring, 4096);
> +	memset(info->ring, 0, vring_size(num, ALIGN));
> +	vring_init(&info->vring, num, info->ring, ALIGN);
>  	info->vq = __vring_new_virtqueue(info->idx, info->vring, vdev, true,
>  					 false, vq_notify, vq_callback, "test");
>  	assert(info->vq);
> @@ -115,7 +118,7 @@ static void vq_info_add(struct vdev_info *dev, int num)
>  	info->idx = dev->nvqs;
>  	info->kick = eventfd(0, EFD_NONBLOCK);
>  	info->call = eventfd(0, EFD_NONBLOCK);
> -	r = posix_memalign(&info->ring, 4096, vring_size(num, 4096));
> +	r = posix_memalign(&info->ring, PAGE_SIZE, vring_size(num, ALIGN));
>  	assert(r >= 0);
>  	vq_reset(info, num, &dev->vdev);
>  	vhost_vq_setup(dev, info);

This is actually doing more than what commit log says.

> @@ -131,7 +134,7 @@ static void vdev_info_init(struct vdev_info* dev, unsigned long long features)
>  	dev->vdev.features = features;
>  	INIT_LIST_HEAD(&dev->vdev.vqs);
>  	spin_lock_init(&dev->vdev.vqs_list_lock);
> -	dev->buf_size = 1024;
> +	dev->buf_size = BUF_SIZE;

This seems to have zero added value.

>  	dev->buf = malloc(dev->buf_size);
>  	assert(dev->buf);
>          dev->control = open("/dev/vhost-test", O_RDWR);
> @@ -396,7 +399,8 @@ int main(int argc, char **argv)
>  
>  done:
>  	vdev_info_init(&dev, features);
> -	vq_info_add(&dev, 256);
> -	run_test(&dev, &dev.vqs[0], delayed, batch, reset, 0x100000);
> +	vq_info_add(&dev, RINGSIZE);
> +
> +	run_test(&dev, &dev.vqs[0], delayed, batch, reset, TEST_BUF_NUM);
>  	return 0;

This replacement does not buy us anything either.

>  }
> -- 
> 2.17.1