Re: [dpdk-dev] [PATCH 4/9] eal: replace usage of blacklist/whitelist in enum

["Gaëtan Rivet" <grive@u256.net>]

Hi,

On 07/06/20 10:01 -0700, Stephen Hemminger wrote:
> This patch renames the enum values in the EAL include files.
> As a backward compatiable temporary migration tool, define
> a replacement mapping for old values.
> 
> The old names relating to blacklist and whitelist are replaced
> by blocklist and allowlist, but applications may be using the
> older compatiablity macros. To help with conversion to new names
> cause a message when the compatabilty names are used.
> 
> This will be upgraded to a warning in the future.
> 
> Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
> ---
>  lib/librte_eal/common/eal_common_devargs.c | 14 +++++++-------
>  lib/librte_eal/include/rte_bus.h           | 12 ++++++++++--
>  lib/librte_eal/include/rte_dev.h           | 12 ++++++++++--
>  lib/librte_eal/include/rte_devargs.h       | 12 ++++++++++--
>  4 files changed, 37 insertions(+), 13 deletions(-)
> 
> diff --git a/lib/librte_eal/common/eal_common_devargs.c b/lib/librte_eal/common/eal_common_devargs.c
> index 2123773ef840..f5e09df267f2 100644
> --- a/lib/librte_eal/common/eal_common_devargs.c
> +++ b/lib/librte_eal/common/eal_common_devargs.c
> @@ -296,7 +296,7 @@ rte_devargs_insert(struct rte_devargs **da)
>  	return 0;
>  }
>  
> -/* store a whitelist parameter for later parsing */
> +/* store a allowlist parameter for later parsing */

a -> an

>  int
>  rte_devargs_add(enum rte_devtype devtype, const char *devargs_str)
>  {
> @@ -313,13 +313,13 @@ rte_devargs_add(enum rte_devtype devtype, const char *devargs_str)
>  		goto fail;
>  	devargs->type = devtype;
>  	bus = devargs->bus;
> -	if (devargs->type == RTE_DEVTYPE_BLACKLISTED_PCI)
> -		devargs->policy = RTE_DEV_BLACKLISTED;
> +	if (devargs->type == RTE_DEVTYPE_BLOCKLIST_PCI)
> +		devargs->policy = RTE_DEV_BLOCKLIST;
>  	if (bus->conf.scan_mode == RTE_BUS_SCAN_UNDEFINED) {
> -		if (devargs->policy == RTE_DEV_WHITELISTED)
> -			bus->conf.scan_mode = RTE_BUS_SCAN_WHITELIST;
> -		else if (devargs->policy == RTE_DEV_BLACKLISTED)
> -			bus->conf.scan_mode = RTE_BUS_SCAN_BLACKLIST;
> +		if (devargs->policy == RTE_DEV_ALLOWLIST)
> +			bus->conf.scan_mode = RTE_BUS_SCAN_ALLOWLIST;
> +		else if (devargs->policy == RTE_DEV_BLOCKLIST)
> +			bus->conf.scan_mode = RTE_BUS_SCAN_BLOCKLIST;
>  	}
>  	TAILQ_INSERT_TAIL(&devargs_list, devargs, next);
>  	return 0;
> diff --git a/lib/librte_eal/include/rte_bus.h b/lib/librte_eal/include/rte_bus.h
> index d3034d0edf77..fcc18a118f44 100644
> --- a/lib/librte_eal/include/rte_bus.h
> +++ b/lib/librte_eal/include/rte_bus.h
> @@ -215,10 +215,18 @@ typedef int (*rte_bus_sigbus_handler_t)(const void *failure_addr);
>   */
>  enum rte_bus_scan_mode {
>  	RTE_BUS_SCAN_UNDEFINED,
> -	RTE_BUS_SCAN_WHITELIST,
> -	RTE_BUS_SCAN_BLACKLIST,
> +	RTE_BUS_SCAN_ALLOWLIST,
> +	RTE_BUS_SCAN_BLOCKLIST,
>  };
>  
> +#define RTE_BUS_SCAN_WHITELIST  \
> +	_Pragma("GCC message \"RTE_BUS_SCAN_WHITELIST\" is deprecated\"") \
> +	RTE_BUS_SCAN_ALLOWLIST
> +
> +#define RTE_BUS_SCAN_BLACKLIST	\
> +	_Pragma("GCC message \"RTE_BUS_SCAN_BLACKLIST\" is deprecated\"") \
> +	RTE_BUS_SCAN_BLOCKLIST
> +
>  /**
>   * A structure used to configure bus operations.
>   */
> diff --git a/lib/librte_eal/include/rte_dev.h b/lib/librte_eal/include/rte_dev.h
> index c8d985fb5cf9..e9184c4a98c1 100644
> --- a/lib/librte_eal/include/rte_dev.h
> +++ b/lib/librte_eal/include/rte_dev.h
> @@ -70,10 +70,18 @@ enum rte_kernel_driver {
>   * Device policies.
>   */
>  enum rte_dev_policy {
> -	RTE_DEV_WHITELISTED,
> -	RTE_DEV_BLACKLISTED,
> +	RTE_DEV_ALLOWLIST,
> +	RTE_DEV_BLOCKLIST,

BLACKLISTED and WHITELISTED were a little contrived to begin with.
The enum was intended to describe the device status.

I think it would be more readable to use RTE_DEV_{BLOCKED,ALLOWED}.

RTE_BUS_SCAN_{BLOCKLIST,ALLOWLIST} is fine.

>  };
>  
> +#define RTE_DEV_WHITELISTED \
> +	_Pragma("GCC message \"'RTE_DEV_WHITELISTED' is deprecated\"") \
> +	RTE_DEV_ALLOWLIST
> +
> +#define RTE_DEV_BLACKLISTED \
> +	_Pragma("GCC message \"'RTE_DEV_BLACKLISTED' is deprecated\"") \
> +	RTE_DEV_BLOCKLIST
> +
>  /**
>   * A generic memory resource representation.
>   */
> diff --git a/lib/librte_eal/include/rte_devargs.h b/lib/librte_eal/include/rte_devargs.h
> index 898efa0d667b..a7c7e1e7962f 100644
> --- a/lib/librte_eal/include/rte_devargs.h
> +++ b/lib/librte_eal/include/rte_devargs.h
> @@ -29,11 +29,19 @@ extern "C" {
>   * Type of generic device
>   */
>  enum rte_devtype {
> -	RTE_DEVTYPE_WHITELISTED_PCI,
> -	RTE_DEVTYPE_BLACKLISTED_PCI,
> +	RTE_DEVTYPE_ALLOWLIST_PCI,
> +	RTE_DEVTYPE_BLOCKLIST_PCI,
>  	RTE_DEVTYPE_VIRTUAL,

I'm ignoring this part, naming will not be aligned on the dev policy,
but this enum is meant to disappear anyway.

>  };
>  
> +#define RTE_DEVTYPE_WHITELISTED_PCI \
> +	_Pragma("GCC message \"'RTE_DEVTYPE_WHITELISTED' is deprecated\"") \
> +	RTE_DEVTYPE_ALLOWLIST_PCI
> +
> +#define RTE_DEVTYPE_BLACKLISTED_PCI \
> +	_Pragma("GCC message \"'RTE_DEVTYPE_BLACKLISTED' is deprecated\"") \
> +	RTE_DEVTYPE_BLOCKLIST_PCI
> +
>  /**
>   * Structure that stores a device given by the user with its arguments
>   *
> -- 
> 2.26.2
> 

-- 
Gaëtan