netdev - [PATCH net-next 11/12] nfp: allow apps to disable ctrl vNIC capabilities

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

Date:   Wed, 17 Jan 2018 18:51:05 -0800
From:   Jakub Kicinski <jakub.kicinski@...ronome.com>
To:     davem@...emloft.net
Cc:     netdev@...r.kernel.org, oss-drivers@...ronome.com,
        Jakub Kicinski <jakub.kicinski@...ronome.com>
Subject: [PATCH net-next 11/12] nfp: allow apps to disable ctrl vNIC capabilities

Most vNIC capabilities are netdev related.  It makes no sense
to initialize them and waste FW resources.  Some are even
counter-productive, like IRQ moderation, which will slow
down exchange of control messages.

Add to nfp_app a mask of enabled control vNIC capabilities
for apps to use.  Make flower and BPF enable all capabilities
for now.  No functional changes.

Signed-off-by: Jakub Kicinski <jakub.kicinski@...ronome.com>
Reviewed-by: Dirk van der Merwe <dirk.vandermerwe@...ronome.com>
---
 drivers/net/ethernet/netronome/nfp/bpf/main.c       | 2 ++
 drivers/net/ethernet/netronome/nfp/flower/main.c    | 2 ++
 drivers/net/ethernet/netronome/nfp/nfp_app.h        | 4 ++++
 drivers/net/ethernet/netronome/nfp/nfp_net_common.c | 4 ++++
 4 files changed, 12 insertions(+)

diff --git a/drivers/net/ethernet/netronome/nfp/bpf/main.c b/drivers/net/ethernet/netronome/nfp/bpf/main.c
index 8823c8360047..5f021d0c88a4 100644
--- a/drivers/net/ethernet/netronome/nfp/bpf/main.c
+++ b/drivers/net/ethernet/netronome/nfp/bpf/main.c
@@ -389,6 +389,8 @@ const struct nfp_app_type app_bpf = {
 	.id		= NFP_APP_BPF_NIC,
 	.name		= "ebpf",
 
+	.ctrl_cap_mask	= ~0U,
+
 	.init		= nfp_bpf_init,
 	.clean		= nfp_bpf_clean,
 
diff --git a/drivers/net/ethernet/netronome/nfp/flower/main.c b/drivers/net/ethernet/netronome/nfp/flower/main.c
index 3c05bffff637..742d6f1575b5 100644
--- a/drivers/net/ethernet/netronome/nfp/flower/main.c
+++ b/drivers/net/ethernet/netronome/nfp/flower/main.c
@@ -565,6 +565,8 @@ static void nfp_flower_stop(struct nfp_app *app)
 const struct nfp_app_type app_flower = {
 	.id		= NFP_APP_FLOWER_NIC,
 	.name		= "flower",
+
+	.ctrl_cap_mask	= ~0U,
 	.ctrl_has_meta	= true,
 
 	.extra_cap	= nfp_flower_extra_cap,
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_app.h b/drivers/net/ethernet/netronome/nfp/nfp_app.h
index 91d469a8e3e6..7e474df90598 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_app.h
+++ b/drivers/net/ethernet/netronome/nfp/nfp_app.h
@@ -66,6 +66,9 @@ extern const struct nfp_app_type app_flower;
  * struct nfp_app_type - application definition
  * @id:		application ID
  * @name:	application name
+ * @ctrl_cap_mask:  ctrl vNIC capability mask, allows disabling features like
+ *		    IRQMOD which are on by default but counter-productive for
+ *		    control messages which are often latency-sensitive
  * @ctrl_has_meta:  control messages have prepend of type:5/port:CTRL
  *
  * Callbacks
@@ -100,6 +103,7 @@ struct nfp_app_type {
 	enum nfp_app_id id;
 	const char *name;
 
+	u32 ctrl_cap_mask;
 	bool ctrl_has_meta;
 
 	int (*init)(struct nfp_app *app);
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
index 86a88770181d..cdf52421eaca 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c
@@ -3790,6 +3790,10 @@ static int nfp_net_read_caps(struct nfp_net *nn)
 		nn->dp.rx_offset = NFP_NET_RX_OFFSET;
 	}
 
+	/* For control vNICs mask out the capabilities app doesn't want. */
+	if (!nn->dp.netdev)
+		nn->cap &= nn->app->type->ctrl_cap_mask;
+
 	return 0;
 }
 
-- 
2.15.1