[dpdk-dev] [PATCH] examples/vhost_crypto: fix zero copy

[Zhang, Roy Fan]

Hi Mattias,

Sorry for the late reply. Comments inline.

> -----Original Message-----
> From: Mattias Rönnblom [mailto:mattias.ronnblom at ericsson.com]
> Sent: Tuesday, October 30, 2018 7:38 PM
> To: Zhang, Roy Fan <roy.fan.zhang at intel.com>; dev at dpdk.org
> Cc: maxime.coquelin at redhat.com
> Subject: Re: [dpdk-dev] [PATCH] examples/vhost_crypto: fix zero copy
> 
> On 2018-10-30 15:48, Fan Zhang wrote:
> > This patch fixes the zero copy enable problem for vhost crypto sample
> > application.
> rte_mempool_free() already does a NULL-check (as per libc free()
> convention), and if you are to do a NULL-check it should be an explicit one
> ("!= NULL").
No problem, I can change that. 
> 
> >   		for (j = 0; j < lo->nb_sockets; j++) {
> >   			rte_vhost_driver_unregister(lo->socket_files[i]);
> > @@ -493,6 +499,19 @@ main(int argc, char *argv[])
> >   		info->nb_vids = lo->nb_sockets;
> >
> >   		rte_cryptodev_info_get(info->cid, &dev_info);
> > +		if (options.zero_copy ==
> RTE_VHOST_CRYPTO_ZERO_COPY_ENABLE) {
> > +#define VHOST_CRYPTO_CDEV_NAME_AESNI_MB_PMD
> 	crypto_aesni_mb
> > +#define VHOST_CRYPTO_CDEV_NAME_AESNI_GCM_PMD
> 	crypto_aesni_gcm
> 
> What's the purpose of these defines?
These two PMD names are defined inside the PMDs and are not exposed to the vhost library.
Also these two PMDs are the ones that copy to the end of the packet for computation purposes, which will cause the problem for virtio case.
 
> 
> > +			if (strstr(dev_info.driver_name,
> > +
> 	RTE_STR(VHOST_CRYPTO_CDEV_NAME_AESNI_MB_PMD)) ||
> > +				strstr(dev_info.driver_name,
> > +
> 	RTE_STR(VHOST_CRYPTO_CDEV_NAME_AESNI_GCM_PMD)))
> > +			RTE_LOG(ERR, USER1, "Cannot enable Zero Copy
> to %s\n",
> > +					dev_info.driver_name);
> 
> "Zero Copy to" should probably be "zero-copy in" or "Zero-copy in".
Thanks I will change that.

Regards,
Fan