* Re: [net-next PATCH 2/3] net/qdisc: IFF_NO_QUEUE drivers should use consistent TX queue len
From: Krister Johansen @ 2016-11-03 20:54 UTC (permalink / raw)
To: Jesper Dangaard Brouer
Cc: netdev, Phil Sutter, Robert Olsson, Jamal Hadi Salim
In-Reply-To: <20161103135606.28737.67383.stgit@firesoul>
On Thu, Nov 03, 2016 at 02:56:06PM +0100, Jesper Dangaard Brouer wrote:
> The flag IFF_NO_QUEUE marks virtual device drivers that doesn't need a
> default qdisc attached, given they will be backed by physical device,
> that already have a qdisc attached for pushback.
>
> It is still supported to attach a qdisc to a IFF_NO_QUEUE device, as
> this can be useful for difference policy reasons (e.g. bandwidth
> limiting containers). For this to work, the tx_queue_len need to have
> a sane value, because some qdiscs inherit/copy the tx_queue_len
> (namely, pfifo, bfifo, gred, htb, plug and sfb).
>
> Commit a813104d9233 ("IFF_NO_QUEUE: Fix for drivers not calling
> ether_setup()") caught situations where some drivers didn't initialize
> tx_queue_len. The problem with the commit was choosing 1 as the
> fallback value.
>
> A qdisc queue length of 1 causes more harm than good, because it
> creates hard to debug situations for userspace. It gives userspace a
> false sense of a working config after attaching a qdisc. As low
> volume traffic (that doesn't activate the qdisc policy) works,
> like ping, while traffic that e.g. needs shaping cannot reach the
> configured policy levels, given the queue length is too small.
Thanks for fixing this. I've run into this in the exact scenario you
describe -- bandwith limiting containers. I'm pretty sure my vote
doesn't count, but I'm in favor of this change.
-K
^ permalink raw reply
* Re: [PATCH net v3] ipv4: allow local fragmentation in ip_finish_output_gso()
From: David Miller @ 2016-11-03 20:56 UTC (permalink / raw)
To: shmulik.ladkani; +Cc: lrichard, netdev, fw, jtluka, hannes
In-Reply-To: <20161103224052.5e066584@halley>
From: Shmulik Ladkani <shmulik.ladkani@gmail.com>
Date: Thu, 3 Nov 2016 22:40:52 +0200
> On Thu, 03 Nov 2016 16:12:44 -0400 (EDT) David Miller <davem@davemloft.net> wrote:
>> Applied and queued up for -stable.
>
> Dave, my response lagged your "Applied" by few minutes ;)
>
> This seems to deserve some more thought to make sure nothing got broken,
> as expressed last in https://patchwork.ozlabs.org/patch/690594/
Feel free to send a followup or a revert if necessary.
^ permalink raw reply
* Re: Time to revisit LISP?
From: Tom Herbert @ 2016-11-03 20:57 UTC (permalink / raw)
To: David Miller; +Cc: chris, Linux Kernel Network Developers
In-Reply-To: <20161103.163758.1592587767905993803.davem@davemloft.net>
On Thu, Nov 3, 2016 at 1:37 PM, David Miller <davem@davemloft.net> wrote:
> From: Tom Herbert <tom@herbertland.com>
> Date: Thu, 3 Nov 2016 12:22:52 -0700
>
>> For instance, one of the his questions is:
>>
>> "What is to keep one from having to service a full Map-Request -->
>> Map-Reply cycle for every packet received?"
>>
>> This can be solved by judicious rate limiting, for instance the
>> infrastructure I implemented to rate limit ILA resolver request could
>> be applied here.
>
> All of these things work great if you have tables that are either very
> tiny or change infrequently.
>
> But once you run into anything seriously dynamic, it has the same
> problems that the routing cache had and OVS can have.
>
One way or another we are going to have to deal with this. If we want
Linux to be serve as router for mobility it is going to have to scale
for having bunches of host routes and they will be quite dynamic
because of mobility.
> And for this reason things like the flow cache are on the chopping
> block. And frankly, I'd remove OVS from the kernel if I could.
>
Quite frankly the DOS issues with OVS were obvious just by looking at
the initial design-- it should have been done better. Yes, if you
upcall every unresolved packet to userspace you're just inviting a DOS
attack.
> Userspace resolution of paths in response to data path signalling
> simply does not scale and is fundamentally an extremely poor design
> choice. We're trying to move away from, rather than towards, these
> kinds of architectures.
OVS is quite different I think. LISP is a specific resolution protocol
of identifier to locator as opposed to be some open ended mechanism to
resolve some arbitrary definition of flows like OVS. Also, I don't
think there's any specific requirement in LISP that prevents on from
implementing the mapping protocol in the kernel, it should just be a
simple UDP communication.
Do you see anything in the protocol itself that would be a showstopper?
Tom
^ permalink raw reply
* Re: Time to revisit LISP?
From: David Miller @ 2016-11-03 21:00 UTC (permalink / raw)
To: tom; +Cc: chris, netdev
In-Reply-To: <CALx6S34-b0dDL0xOvjU+uDzwsZHAt4QPfpHrpeauMmMh4MTKXA@mail.gmail.com>
From: Tom Herbert <tom@herbertland.com>
Date: Thu, 3 Nov 2016 13:57:59 -0700
> On Thu, Nov 3, 2016 at 1:37 PM, David Miller <davem@davemloft.net> wrote:
>> Userspace resolution of paths in response to data path signalling
>> simply does not scale and is fundamentally an extremely poor design
>> choice. We're trying to move away from, rather than towards, these
>> kinds of architectures.
>
> OVS is quite different I think. LISP is a specific resolution protocol
> of identifier to locator as opposed to be some open ended mechanism to
> resolve some arbitrary definition of flows like OVS. Also, I don't
> think there's any specific requirement in LISP that prevents on from
> implementing the mapping protocol in the kernel, it should just be a
> simple UDP communication.
>
> Do you see anything in the protocol itself that would be a showstopper?
I'd have to see the code and how it works. I can't review hypothetical
implementations.
^ permalink raw reply
* Re: [PATCH net v3] ipv4: allow local fragmentation in ip_finish_output_gso()
From: Lance Richardson @ 2016-11-03 21:05 UTC (permalink / raw)
To: Shmulik Ladkani; +Cc: fw, hannes, netdev, jtluka
In-Reply-To: <20161103222751.5ae120ed@halley>
> From: "Shmulik Ladkani" <shmulik.ladkani@gmail.com>
> To: "Lance Richardson" <lrichard@redhat.com>, fw@strlen.de, hannes@stressinduktion.org
> Cc: netdev@vger.kernel.org, jtluka@redhat.com
> Sent: Thursday, November 3, 2016 4:27:51 PM
> Subject: Re: [PATCH net v3] ipv4: allow local fragmentation in ip_finish_output_gso()
>
> Hi Hannes, Lance,
>
> On Wed, 2 Nov 2016 16:36:17 -0400 Lance Richardson <lrichard@redhat.com>
> wrote:
> >
> > - if (skb_iif && !(df & htons(IP_DF))) {
> > - /* Arrived from an ingress interface, got encapsulated, with
> > - * fragmentation of encapulating frames allowed.
> > - * If skb is gso, the resulting encapsulated network segments
> > - * may exceed dst mtu.
> > - * Allow IP Fragmentation of segments.
> > - */
> > - IPCB(skb)->flags |= IPSKB_FRAG_SEGS;
> > - }
>
> Thinking this over, I'm concerned of this change.
>
> Few months back, we discussed this and got to the conclusion that in the
> "ingress,tunnel,egress" scenario, segments are allowed to be
> fragmented if the original inner ip packet does NOT have the DF.
>
> See
> https://patchwork.ozlabs.org/patch/657132/
> https://patchwork.ozlabs.org/patch/661219/
>
> I think you expressed that those tunneled skbs already having DF set
> should go through pmtu discovery.
>
> Suggested patch unconditionally calls skb_gso_validate_mtu().
>
> Thus we're changing behavior for "ingress,tunnel,egress" scenario of
> the tunneled packets having DF set in the inner iph.
>
> WDYT?
>
I'm still digesting the patchwork history, but it seems to me:
1) If we call skb_gso_validate_mtu() and it returns true, ip_finish_output2() will
be called, just as before, so nothing changes.
2) If we were to avoid calling skb_gso_validate_mtu() when it would have returned
false and call ip_finish_output2() without performing fragmentation, we
would transmit (or attempt to transmit) a packet that exceeds the configured
MTU.
3) If we do call skb_gso_validate_mtu() and it returns false, ip_finish_output_gso()
will call ip_fragment() to perform needed fragmentation. Whether fragmentation
actually occurs at this point depends on the value of the DF flag in the
IP header (and perhaps skb->ignore_df, frag_max_size, etc.).
Is the issue you're pointing out about cases in which the inner IP header has DF set
but the tunnel header does not?
Thanks,
Lance
^ permalink raw reply
* Re: [PATCH net v3] ipv4: allow local fragmentation in ip_finish_output_gso()
From: Hannes Frederic Sowa @ 2016-11-03 21:34 UTC (permalink / raw)
To: Lance Richardson, Shmulik Ladkani; +Cc: fw, netdev, jtluka
In-Reply-To: <1769078966.96766888.1478207154311.JavaMail.zimbra@redhat.com>
On 03.11.2016 22:05, Lance Richardson wrote:
>> From: "Shmulik Ladkani" <shmulik.ladkani@gmail.com>
>> To: "Lance Richardson" <lrichard@redhat.com>, fw@strlen.de, hannes@stressinduktion.org
>> Cc: netdev@vger.kernel.org, jtluka@redhat.com
>> Sent: Thursday, November 3, 2016 4:27:51 PM
>> Subject: Re: [PATCH net v3] ipv4: allow local fragmentation in ip_finish_output_gso()
>>
>> Hi Hannes, Lance,
>>
>> On Wed, 2 Nov 2016 16:36:17 -0400 Lance Richardson <lrichard@redhat.com>
>> wrote:
>>>
>>> - if (skb_iif && !(df & htons(IP_DF))) {
>>> - /* Arrived from an ingress interface, got encapsulated, with
>>> - * fragmentation of encapulating frames allowed.
>>> - * If skb is gso, the resulting encapsulated network segments
>>> - * may exceed dst mtu.
>>> - * Allow IP Fragmentation of segments.
>>> - */
>>> - IPCB(skb)->flags |= IPSKB_FRAG_SEGS;
>>> - }
>>
>> Thinking this over, I'm concerned of this change.
>>
>> Few months back, we discussed this and got to the conclusion that in the
>> "ingress,tunnel,egress" scenario, segments are allowed to be
>> fragmented if the original inner ip packet does NOT have the DF.
>>
>> See
>> https://patchwork.ozlabs.org/patch/657132/
>> https://patchwork.ozlabs.org/patch/661219/
>>
>> I think you expressed that those tunneled skbs already having DF set
>> should go through pmtu discovery.
>>
>> Suggested patch unconditionally calls skb_gso_validate_mtu().
>>
>> Thus we're changing behavior for "ingress,tunnel,egress" scenario of
>> the tunneled packets having DF set in the inner iph.
>>
>> WDYT?
>>
>
> I'm still digesting the patchwork history, but it seems to me:
>
> 1) If we call skb_gso_validate_mtu() and it returns true, ip_finish_output2() will
> be called, just as before, so nothing changes.
>
> 2) If we were to avoid calling skb_gso_validate_mtu() when it would have returned
> false and call ip_finish_output2() without performing fragmentation, we
> would transmit (or attempt to transmit) a packet that exceeds the configured
> MTU.
>
> 3) If we do call skb_gso_validate_mtu() and it returns false, ip_finish_output_gso()
> will call ip_fragment() to perform needed fragmentation. Whether fragmentation
> actually occurs at this point depends on the value of the DF flag in the
> IP header (and perhaps skb->ignore_df, frag_max_size, etc.).
>
> Is the issue you're pointing out about cases in which the inner IP header has DF set
> but the tunnel header does not?
Correct, but we should maybe redefine the code a bit. From my
understanding we can now create an ICMP storm in case every fragment gets.
I think for net this is currently fine and we certainly don't want to
generate oversized datagrams.
I fear the more special case logic we add the sooner or later it will
bite us again. :/
Right now, I see the problem we might end up generating lots of error
callbacks for large gso segments. Maybe we can also just abort after
fragmenting the first frame generated an error. Florian? Or just
overoptimize and jump to the last one, which could have a different size. :)
Anyway, the best thing would be that vxlan etc. inherits the inner DF bit.
The problem to solve here is that for some time we generated oversized
packets on the wire, which is absolutely a no-go. If we now start to
break things for people with "wrong" configuration, we could also get
more complains. Currently I think this is the only way out, unfortunately.
Any other ideas?
^ permalink raw reply
* [PATCH net-next V4 0/9] liquidio CN23XX VF support
From: Raghu Vatsavayi @ 2016-11-03 21:35 UTC (permalink / raw)
To: davem; +Cc: netdev, Raghu Vatsavayi
Dave,
Following is the V4 patch series for adding VF support on
CN23XX devices. This version addressed:
1) Your concern for ordering of local variable declarations
from longest to shortest line.
2) Removed module parameters max_vfs, num_queues_per_{p,v}f.
3) Minor changes for fixing new checkpatch script related
errors on pre-existing driver.
I will post remaining VF patches soon after this patchseries is
applied. Please apply patches in the following order as some of
the patches depend on earlier patches.
Thanks.
Raghu Vatsavayi (9):
liquidio CN23XX: HW config for VF support
liquidio CN23XX: sysfs VF config support
liquidio CN23XX: Mailbox support
liquidio CN23XX: mailbox interrupt processing
liquidio CN23XX: VF related operations
liquidio CN23XX: device states
liquidio CN23XX: code cleanup
liquidio CN23XX: copyrights changes and alignment
liquidio CN23XX: fix for new check patch errors
drivers/net/ethernet/cavium/liquidio/Makefile | 1 +
.../ethernet/cavium/liquidio/cn23xx_pf_device.c | 316 ++++++++++++---
.../ethernet/cavium/liquidio/cn23xx_pf_device.h | 44 +--
.../net/ethernet/cavium/liquidio/cn23xx_pf_regs.h | 51 ++-
.../net/ethernet/cavium/liquidio/cn66xx_device.c | 49 +--
.../net/ethernet/cavium/liquidio/cn66xx_device.h | 41 +-
drivers/net/ethernet/cavium/liquidio/cn66xx_regs.h | 49 ++-
.../net/ethernet/cavium/liquidio/cn68xx_device.c | 38 +-
.../net/ethernet/cavium/liquidio/cn68xx_device.h | 37 +-
drivers/net/ethernet/cavium/liquidio/cn68xx_regs.h | 37 +-
drivers/net/ethernet/cavium/liquidio/lio_core.c | 68 +++-
drivers/net/ethernet/cavium/liquidio/lio_ethtool.c | 65 ++--
drivers/net/ethernet/cavium/liquidio/lio_main.c | 424 ++++++++++++++++++---
.../net/ethernet/cavium/liquidio/liquidio_common.h | 100 +++--
.../net/ethernet/cavium/liquidio/liquidio_image.h | 36 +-
.../net/ethernet/cavium/liquidio/octeon_config.h | 46 ++-
.../net/ethernet/cavium/liquidio/octeon_console.c | 156 ++++----
.../net/ethernet/cavium/liquidio/octeon_device.c | 79 ++--
.../net/ethernet/cavium/liquidio/octeon_device.h | 138 ++++---
drivers/net/ethernet/cavium/liquidio/octeon_droq.c | 91 +++--
drivers/net/ethernet/cavium/liquidio/octeon_droq.h | 18 +-
drivers/net/ethernet/cavium/liquidio/octeon_iq.h | 25 +-
.../net/ethernet/cavium/liquidio/octeon_mailbox.c | 318 ++++++++++++++++
.../net/ethernet/cavium/liquidio/octeon_mailbox.h | 115 ++++++
drivers/net/ethernet/cavium/liquidio/octeon_main.h | 47 +--
.../net/ethernet/cavium/liquidio/octeon_mem_ops.c | 7 +-
.../net/ethernet/cavium/liquidio/octeon_mem_ops.h | 5 +-
.../net/ethernet/cavium/liquidio/octeon_network.h | 11 +-
drivers/net/ethernet/cavium/liquidio/octeon_nic.c | 5 +-
drivers/net/ethernet/cavium/liquidio/octeon_nic.h | 7 +-
.../net/ethernet/cavium/liquidio/request_manager.c | 34 +-
.../ethernet/cavium/liquidio/response_manager.c | 11 +-
.../ethernet/cavium/liquidio/response_manager.h | 6 +-
33 files changed, 1677 insertions(+), 798 deletions(-)
create mode 100644 drivers/net/ethernet/cavium/liquidio/octeon_mailbox.c
create mode 100644 drivers/net/ethernet/cavium/liquidio/octeon_mailbox.h
--
1.8.3.1
^ permalink raw reply
* [PATCH net-next V4 1/9] liquidio CN23XX: HW config for VF support
From: Raghu Vatsavayi @ 2016-11-03 21:35 UTC (permalink / raw)
To: davem
Cc: netdev, Raghu Vatsavayi, Raghu Vatsavayi, Derek Chickles,
Satanand Burla, Felix Manlunas
In-Reply-To: <1478208918-8330-1-git-send-email-rvatsavayi@caviumnetworks.com>
Adds support for configuring HW for creating VFs.
Signed-off-by: Raghu Vatsavayi <raghu.vatsavayi@caviumnetworks.com>
Signed-off-by: Derek Chickles <derek.chickles@caviumnetworks.com>
Signed-off-by: Satanand Burla <satananda.burla@caviumnetworks.com>
Signed-off-by: Felix Manlunas <felix.manlunas@caviumnetworks.com>
---
.../ethernet/cavium/liquidio/cn23xx_pf_device.c | 84 +++++++++++++---------
.../net/ethernet/cavium/liquidio/octeon_config.h | 6 ++
.../net/ethernet/cavium/liquidio/octeon_device.h | 12 +++-
3 files changed, 68 insertions(+), 34 deletions(-)
diff --git a/drivers/net/ethernet/cavium/liquidio/cn23xx_pf_device.c b/drivers/net/ethernet/cavium/liquidio/cn23xx_pf_device.c
index 380a641..4c93eac 100644
--- a/drivers/net/ethernet/cavium/liquidio/cn23xx_pf_device.c
+++ b/drivers/net/ethernet/cavium/liquidio/cn23xx_pf_device.c
@@ -40,11 +40,6 @@
*/
#define CN23XX_INPUT_JABBER 64600
-#define LIOLUT_RING_DISTRIBUTION 9
-const int liolut_num_vfs_to_rings_per_vf[LIOLUT_RING_DISTRIBUTION] = {
- 0, 8, 4, 2, 2, 2, 1, 1, 1
-};
-
void cn23xx_dump_pf_initialized_regs(struct octeon_device *oct)
{
int i = 0;
@@ -309,9 +304,10 @@ u32 cn23xx_pf_get_oq_ticks(struct octeon_device *oct, u32 time_intr_in_us)
static void cn23xx_setup_global_mac_regs(struct octeon_device *oct)
{
- u64 reg_val;
u16 mac_no = oct->pcie_port;
u16 pf_num = oct->pf_num;
+ u64 reg_val;
+ u64 temp;
/* programming SRN and TRS for each MAC(0..3) */
@@ -333,6 +329,14 @@ static void cn23xx_setup_global_mac_regs(struct octeon_device *oct)
/* setting TRS <23:16> */
reg_val = reg_val |
(oct->sriov_info.trs << CN23XX_PKT_MAC_CTL_RINFO_TRS_BIT_POS);
+ /* setting RPVF <39:32> */
+ temp = oct->sriov_info.rings_per_vf & 0xff;
+ reg_val |= (temp << CN23XX_PKT_MAC_CTL_RINFO_RPVF_BIT_POS);
+
+ /* setting NVFS <55:48> */
+ temp = oct->sriov_info.max_vfs & 0xff;
+ reg_val |= (temp << CN23XX_PKT_MAC_CTL_RINFO_NVFS_BIT_POS);
+
/* write these settings to MAC register */
octeon_write_csr64(oct, CN23XX_SLI_PKT_MAC_RINFO64(mac_no, pf_num),
reg_val);
@@ -399,11 +403,12 @@ static int cn23xx_reset_io_queues(struct octeon_device *oct)
static int cn23xx_pf_setup_global_input_regs(struct octeon_device *oct)
{
+ struct octeon_cn23xx_pf *cn23xx = (struct octeon_cn23xx_pf *)oct->chip;
+ struct octeon_instr_queue *iq;
+ u64 intr_threshold, reg_val;
u32 q_no, ern, srn;
u64 pf_num;
- u64 intr_threshold, reg_val;
- struct octeon_instr_queue *iq;
- struct octeon_cn23xx_pf *cn23xx = (struct octeon_cn23xx_pf *)oct->chip;
+ u64 vf_num;
pf_num = oct->pf_num;
@@ -420,6 +425,16 @@ static int cn23xx_pf_setup_global_input_regs(struct octeon_device *oct)
*/
for (q_no = 0; q_no < ern; q_no++) {
reg_val = oct->pcie_port << CN23XX_PKT_INPUT_CTL_MAC_NUM_POS;
+
+ /* for VF assigned queues. */
+ if (q_no < oct->sriov_info.pf_srn) {
+ vf_num = q_no / oct->sriov_info.rings_per_vf;
+ vf_num += 1; /* VF1, VF2,........ */
+ } else {
+ vf_num = 0;
+ }
+
+ reg_val |= vf_num << CN23XX_PKT_INPUT_CTL_VF_NUM_POS;
reg_val |= pf_num << CN23XX_PKT_INPUT_CTL_PF_NUM_POS;
octeon_write_csr64(oct, CN23XX_SLI_IQ_PKT_CONTROL64(q_no),
@@ -1048,50 +1063,53 @@ static void cn23xx_setup_reg_address(struct octeon_device *oct)
static int cn23xx_sriov_config(struct octeon_device *oct)
{
- u32 total_rings;
struct octeon_cn23xx_pf *cn23xx = (struct octeon_cn23xx_pf *)oct->chip;
- /* num_vfs is already filled for us */
+ u32 max_rings, total_rings, max_vfs;
u32 pf_srn, num_pf_rings;
+ u32 max_possible_vfs;
cn23xx->conf =
- (struct octeon_config *)oct_get_config_info(oct, LIO_23XX);
+ (struct octeon_config *)oct_get_config_info(oct, LIO_23XX);
switch (oct->rev_id) {
case OCTEON_CN23XX_REV_1_0:
- total_rings = CN23XX_MAX_RINGS_PER_PF_PASS_1_0;
+ max_rings = CN23XX_MAX_RINGS_PER_PF_PASS_1_0;
+ max_possible_vfs = CN23XX_MAX_VFS_PER_PF_PASS_1_0;
break;
case OCTEON_CN23XX_REV_1_1:
- total_rings = CN23XX_MAX_RINGS_PER_PF_PASS_1_1;
+ max_rings = CN23XX_MAX_RINGS_PER_PF_PASS_1_1;
+ max_possible_vfs = CN23XX_MAX_VFS_PER_PF_PASS_1_1;
break;
default:
- total_rings = CN23XX_MAX_RINGS_PER_PF;
+ max_rings = CN23XX_MAX_RINGS_PER_PF;
+ max_possible_vfs = CN23XX_MAX_VFS_PER_PF;
break;
}
- if (!oct->sriov_info.num_pf_rings) {
- if (total_rings > num_present_cpus())
- num_pf_rings = num_present_cpus();
- else
- num_pf_rings = total_rings;
- } else {
- num_pf_rings = oct->sriov_info.num_pf_rings;
- if (num_pf_rings > total_rings) {
- dev_warn(&oct->pci_dev->dev,
- "num_queues_per_pf requested %u is more than available rings. Reducing to %u\n",
- num_pf_rings, total_rings);
- num_pf_rings = total_rings;
- }
- }
+ if (max_rings <= num_present_cpus())
+ num_pf_rings = 1;
+ else
+ num_pf_rings = num_present_cpus();
+
+ max_vfs = min_t(u32,
+ (max_rings - num_pf_rings), max_possible_vfs);
+
+ total_rings = num_pf_rings + max_vfs;
- total_rings = num_pf_rings;
/* the first ring of the pf */
pf_srn = total_rings - num_pf_rings;
oct->sriov_info.trs = total_rings;
+ oct->sriov_info.max_vfs = max_vfs;
+ oct->sriov_info.rings_per_vf = 1;
oct->sriov_info.pf_srn = pf_srn;
oct->sriov_info.num_pf_rings = num_pf_rings;
- dev_dbg(&oct->pci_dev->dev, "trs:%d pf_srn:%d num_pf_rings:%d\n",
- oct->sriov_info.trs, oct->sriov_info.pf_srn,
- oct->sriov_info.num_pf_rings);
+ dev_notice(&oct->pci_dev->dev, "trs:%d max_vfs:%d rings_per_vf:%d pf_srn:%d num_pf_rings:%d\n",
+ oct->sriov_info.trs, oct->sriov_info.max_vfs,
+ oct->sriov_info.rings_per_vf, oct->sriov_info.pf_srn,
+ oct->sriov_info.num_pf_rings);
+
+ oct->sriov_info.sriov_enabled = 0;
+
return 0;
}
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_config.h b/drivers/net/ethernet/cavium/liquidio/octeon_config.h
index c765568..512bca5 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_config.h
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_config.h
@@ -65,9 +65,15 @@
#define DEFAULT_NUM_NIC_PORTS_68XX_210NV 2
/* CN23xx IQ configuration macros */
+#define CN23XX_MAX_VFS_PER_PF_PASS_1_0 8
+#define CN23XX_MAX_VFS_PER_PF_PASS_1_1 31
+#define CN23XX_MAX_VFS_PER_PF 63
+#define CN23XX_MAX_RINGS_PER_VF 8
+
#define CN23XX_MAX_RINGS_PER_PF_PASS_1_0 12
#define CN23XX_MAX_RINGS_PER_PF_PASS_1_1 32
#define CN23XX_MAX_RINGS_PER_PF 64
+#define CN23XX_MAX_RINGS_PER_VF 8
#define CN23XX_MAX_INPUT_QUEUES CN23XX_MAX_RINGS_PER_PF
#define CN23XX_MAX_IQ_DESCRIPTORS 2048
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_device.h b/drivers/net/ethernet/cavium/liquidio/octeon_device.h
index da15c2a..751d3b6 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_device.h
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_device.h
@@ -322,11 +322,21 @@ struct octeon_pf_vf_hs_word {
};
struct octeon_sriov_info {
+ /* Number of rings assigned to VF */
+ u32 rings_per_vf;
+
+ /** Max Number of VF devices that can be enabled. This variable can
+ * specified during load time or it will be derived after allocating
+ * PF queues. When max_vfs is derived then each VF will get one queue
+ **/
+ u32 max_vfs;
+
/* Actual rings left for PF device */
u32 num_pf_rings;
- /* SRN of PF usable IO queues */
+ /* SRN of PF usable IO queues */
u32 pf_srn;
+
/* total pf rings */
u32 trs;
--
1.8.3.1
^ permalink raw reply related
* [PATCH net-next V4 2/9] liquidio CN23XX: sysfs VF config support
From: Raghu Vatsavayi @ 2016-11-03 21:35 UTC (permalink / raw)
To: davem
Cc: netdev, Raghu Vatsavayi, Raghu Vatsavayi, Derek Chickles,
Satanand Burla, Felix Manlunas
In-Reply-To: <1478208918-8330-1-git-send-email-rvatsavayi@caviumnetworks.com>
Adds sysfs based support for enabling or disabling VFs.
Signed-off-by: Raghu Vatsavayi <raghu.vatsavayi@caviumnetworks.com>
Signed-off-by: Derek Chickles <derek.chickles@caviumnetworks.com>
Signed-off-by: Satanand Burla <satananda.burla@caviumnetworks.com>
Signed-off-by: Felix Manlunas <felix.manlunas@caviumnetworks.com>
---
drivers/net/ethernet/cavium/liquidio/lio_main.c | 98 ++++++++++++++++++++++
.../net/ethernet/cavium/liquidio/octeon_config.h | 3 +
.../net/ethernet/cavium/liquidio/octeon_device.h | 8 ++
3 files changed, 109 insertions(+)
diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c
index 71d01a7..29f5985 100644
--- a/drivers/net/ethernet/cavium/liquidio/lio_main.c
+++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c
@@ -180,6 +180,8 @@ struct octeon_device_priv {
unsigned long napi_mask;
};
+static int liquidio_enable_sriov(struct pci_dev *dev, int num_vfs);
+
static int octeon_device_init(struct octeon_device *);
static int liquidio_stop(struct net_device *netdev);
static void liquidio_remove(struct pci_dev *pdev);
@@ -518,6 +520,7 @@ static int liquidio_resume(struct pci_dev *pdev __attribute__((unused)))
.suspend = liquidio_suspend,
.resume = liquidio_resume,
#endif
+ .sriov_configure = liquidio_enable_sriov,
};
/**
@@ -1472,6 +1475,8 @@ static void octeon_destroy_resources(struct octeon_device *oct)
continue;
octeon_delete_instr_queue(oct, i);
}
+ if (oct->sriov_info.sriov_enabled)
+ pci_disable_sriov(oct->pci_dev);
/* fallthrough */
case OCT_DEV_SC_BUFF_POOL_INIT_DONE:
octeon_free_sc_buffer_pool(oct);
@@ -3990,6 +3995,99 @@ static int setup_nic_devices(struct octeon_device *octeon_dev)
return -ENODEV;
}
+static int octeon_enable_sriov(struct octeon_device *oct)
+{
+ unsigned int num_vfs_alloced = oct->sriov_info.num_vfs_alloced;
+ struct pci_dev *vfdev;
+ int err;
+ u32 u;
+
+ if (OCTEON_CN23XX_PF(oct) && num_vfs_alloced) {
+ err = pci_enable_sriov(oct->pci_dev,
+ oct->sriov_info.num_vfs_alloced);
+ if (err) {
+ dev_err(&oct->pci_dev->dev,
+ "OCTEON: Failed to enable PCI sriov: %d\n",
+ err);
+ oct->sriov_info.num_vfs_alloced = 0;
+ return err;
+ }
+ oct->sriov_info.sriov_enabled = 1;
+
+ /* init lookup table that maps DPI ring number to VF pci_dev
+ * struct pointer
+ */
+ u = 0;
+ vfdev = pci_get_device(PCI_VENDOR_ID_CAVIUM,
+ OCTEON_CN23XX_VF_VID, NULL);
+ while (vfdev) {
+ if (vfdev->is_virtfn &&
+ (vfdev->physfn == oct->pci_dev)) {
+ oct->sriov_info.dpiring_to_vfpcidev_lut[u] =
+ vfdev;
+ u += oct->sriov_info.rings_per_vf;
+ }
+ vfdev = pci_get_device(PCI_VENDOR_ID_CAVIUM,
+ OCTEON_CN23XX_VF_VID, vfdev);
+ }
+ }
+
+ return num_vfs_alloced;
+}
+
+static int lio_pci_sriov_disable(struct octeon_device *oct)
+{
+ int u;
+
+ if (pci_vfs_assigned(oct->pci_dev)) {
+ dev_err(&oct->pci_dev->dev, "VFs are still assigned to VMs.\n");
+ return -EPERM;
+ }
+
+ pci_disable_sriov(oct->pci_dev);
+
+ u = 0;
+ while (u < MAX_POSSIBLE_VFS) {
+ oct->sriov_info.dpiring_to_vfpcidev_lut[u] = NULL;
+ u += oct->sriov_info.rings_per_vf;
+ }
+
+ oct->sriov_info.num_vfs_alloced = 0;
+ dev_info(&oct->pci_dev->dev, "oct->pf_num:%d disabled VFs\n",
+ oct->pf_num);
+
+ return 0;
+}
+
+static int liquidio_enable_sriov(struct pci_dev *dev, int num_vfs)
+{
+ struct octeon_device *oct = pci_get_drvdata(dev);
+ int ret = 0;
+
+ if ((num_vfs == oct->sriov_info.num_vfs_alloced) &&
+ (oct->sriov_info.sriov_enabled)) {
+ dev_info(&oct->pci_dev->dev, "oct->pf_num:%d already enabled num_vfs:%d\n",
+ oct->pf_num, num_vfs);
+ return 0;
+ }
+
+ if (!num_vfs) {
+ ret = lio_pci_sriov_disable(oct);
+ } else if (num_vfs > oct->sriov_info.max_vfs) {
+ dev_err(&oct->pci_dev->dev,
+ "OCTEON: Max allowed VFs:%d user requested:%d",
+ oct->sriov_info.max_vfs, num_vfs);
+ ret = -EPERM;
+ } else {
+ oct->sriov_info.num_vfs_alloced = num_vfs;
+ ret = octeon_enable_sriov(oct);
+ dev_info(&oct->pci_dev->dev, "oct->pf_num:%d enabled num_vfs:%d\n",
+ oct->pf_num, num_vfs);
+ }
+
+ return ret;
+}
+
/**
* \brief initialize the NIC
* @param oct octeon device
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_config.h b/drivers/net/ethernet/cavium/liquidio/octeon_config.h
index 512bca5..71daf82 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_config.h
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_config.h
@@ -472,4 +472,7 @@ struct octeon_config {
#define MAX_POSSIBLE_OCTEON_INSTR_QUEUES CN23XX_MAX_INPUT_QUEUES
#define MAX_POSSIBLE_OCTEON_OUTPUT_QUEUES CN23XX_MAX_OUTPUT_QUEUES
+
+#define MAX_POSSIBLE_VFS 64
+
#endif /* __OCTEON_CONFIG_H__ */
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_device.h b/drivers/net/ethernet/cavium/liquidio/octeon_device.h
index 751d3b6..cfd12ec 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_device.h
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_device.h
@@ -38,6 +38,7 @@
#define OCTEON_CN68XX 0x0091
#define OCTEON_CN66XX 0x0092
#define OCTEON_CN23XX_PF_VID 0x9702
+#define OCTEON_CN23XX_VF_VID 0x9712
/**RevisionId for the chips */
#define OCTEON_CN23XX_REV_1_0 0x00
@@ -331,6 +332,9 @@ struct octeon_sriov_info {
**/
u32 max_vfs;
+ /** Number of VF devices enabled using sysfs. */
+ u32 num_vfs_alloced;
+
/* Actual rings left for PF device */
u32 num_pf_rings;
@@ -340,6 +344,10 @@ struct octeon_sriov_info {
/* total pf rings */
u32 trs;
+ u32 sriov_enabled;
+
+ /*lookup table that maps DPI ring number to VF pci_dev struct pointer*/
+ struct pci_dev *dpiring_to_vfpcidev_lut[MAX_POSSIBLE_VFS];
};
struct octeon_ioq_vector {
--
1.8.3.1
^ permalink raw reply related
* [PATCH net-next V4 3/9] liquidio CN23XX: Mailbox support
From: Raghu Vatsavayi @ 2016-11-03 21:35 UTC (permalink / raw)
To: davem
Cc: netdev, Raghu Vatsavayi, Raghu Vatsavayi, Derek Chickles,
Satanand Burla, Felix Manlunas
In-Reply-To: <1478208918-8330-1-git-send-email-rvatsavayi@caviumnetworks.com>
Adds support for mailbox communication between PF and VF.
Signed-off-by: Raghu Vatsavayi <raghu.vatsavayi@caviumnetworks.com>
Signed-off-by: Derek Chickles <derek.chickles@caviumnetworks.com>
Signed-off-by: Satanand Burla <satananda.burla@caviumnetworks.com>
Signed-off-by: Felix Manlunas <felix.manlunas@caviumnetworks.com>
---
drivers/net/ethernet/cavium/liquidio/Makefile | 1 +
drivers/net/ethernet/cavium/liquidio/lio_core.c | 32 +++
.../net/ethernet/cavium/liquidio/liquidio_common.h | 6 +-
.../net/ethernet/cavium/liquidio/octeon_device.h | 4 +
.../net/ethernet/cavium/liquidio/octeon_mailbox.c | 318 +++++++++++++++++++++
.../net/ethernet/cavium/liquidio/octeon_mailbox.h | 115 ++++++++
drivers/net/ethernet/cavium/liquidio/octeon_main.h | 2 +-
7 files changed, 475 insertions(+), 3 deletions(-)
create mode 100644 drivers/net/ethernet/cavium/liquidio/octeon_mailbox.c
create mode 100644 drivers/net/ethernet/cavium/liquidio/octeon_mailbox.h
diff --git a/drivers/net/ethernet/cavium/liquidio/Makefile b/drivers/net/ethernet/cavium/liquidio/Makefile
index 5a27b2a..14958de 100644
--- a/drivers/net/ethernet/cavium/liquidio/Makefile
+++ b/drivers/net/ethernet/cavium/liquidio/Makefile
@@ -11,6 +11,7 @@ liquidio-$(CONFIG_LIQUIDIO) += lio_ethtool.o \
cn66xx_device.o \
cn68xx_device.o \
cn23xx_pf_device.o \
+ octeon_mailbox.o \
octeon_mem_ops.o \
octeon_droq.o \
octeon_nic.o
diff --git a/drivers/net/ethernet/cavium/liquidio/lio_core.c b/drivers/net/ethernet/cavium/liquidio/lio_core.c
index 201eddb..e6026df 100644
--- a/drivers/net/ethernet/cavium/liquidio/lio_core.c
+++ b/drivers/net/ethernet/cavium/liquidio/lio_core.c
@@ -264,3 +264,35 @@ void liquidio_link_ctrl_cmd_completion(void *nctrl_ptr)
nctrl->ncmd.s.cmd);
}
}
+
+void octeon_pf_changed_vf_macaddr(struct octeon_device *oct, u8 *mac)
+{
+ bool macaddr_changed = false;
+ struct net_device *netdev;
+ struct lio *lio;
+
+ rtnl_lock();
+
+ netdev = oct->props[0].netdev;
+ lio = GET_LIO(netdev);
+
+ lio->linfo.macaddr_is_admin_asgnd = true;
+
+ if (!ether_addr_equal(netdev->dev_addr, mac)) {
+ macaddr_changed = true;
+ ether_addr_copy(netdev->dev_addr, mac);
+ ether_addr_copy(((u8 *)&lio->linfo.hw_addr) + 2, mac);
+ call_netdevice_notifiers(NETDEV_CHANGEADDR, netdev);
+ }
+
+ rtnl_unlock();
+
+ if (macaddr_changed)
+ dev_info(&oct->pci_dev->dev,
+ "PF changed VF's MAC address to %02hhx:%02hhx:%02hhx:%02hhx:%02hhx:%02hhx\n",
+ mac[0], mac[1], mac[2], mac[3], mac[4], mac[5]);
+
+ /* no need to notify the firmware of the macaddr change because
+ * the PF did that already
+ */
+}
diff --git a/drivers/net/ethernet/cavium/liquidio/liquidio_common.h b/drivers/net/ethernet/cavium/liquidio/liquidio_common.h
index 0d990ac..caeff9a 100644
--- a/drivers/net/ethernet/cavium/liquidio/liquidio_common.h
+++ b/drivers/net/ethernet/cavium/liquidio/liquidio_common.h
@@ -731,13 +731,15 @@ struct oct_link_info {
#ifdef __BIG_ENDIAN_BITFIELD
u64 gmxport:16;
- u64 rsvd:32;
+ u64 macaddr_is_admin_asgnd:1;
+ u64 rsvd:31;
u64 num_txpciq:8;
u64 num_rxpciq:8;
#else
u64 num_rxpciq:8;
u64 num_txpciq:8;
- u64 rsvd:32;
+ u64 rsvd:31;
+ u64 macaddr_is_admin_asgnd:1;
u64 gmxport:16;
#endif
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_device.h b/drivers/net/ethernet/cavium/liquidio/octeon_device.h
index cfd12ec..77a6eb7 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_device.h
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_device.h
@@ -492,6 +492,9 @@ struct octeon_device {
int msix_on;
+ /** Mail Box details of each octeon queue. */
+ struct octeon_mbox *mbox[MAX_POSSIBLE_VFS];
+
/** IOq information of it's corresponding MSI-X interrupt. */
struct octeon_ioq_vector *ioq_vector;
@@ -511,6 +514,7 @@ struct octeon_device {
#define OCTEON_CN6XXX(oct) ((oct->chip_id == OCTEON_CN66XX) || \
(oct->chip_id == OCTEON_CN68XX))
#define OCTEON_CN23XX_PF(oct) (oct->chip_id == OCTEON_CN23XX_PF_VID)
+#define OCTEON_CN23XX_VF(oct) ((oct)->chip_id == OCTEON_CN23XX_VF_VID)
#define CHIP_FIELD(oct, TYPE, field) \
(((struct octeon_ ## TYPE *)(oct->chip))->field)
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_mailbox.c b/drivers/net/ethernet/cavium/liquidio/octeon_mailbox.c
new file mode 100644
index 0000000..5309384
--- /dev/null
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_mailbox.c
@@ -0,0 +1,318 @@
+/**********************************************************************
+ * Author: Cavium, Inc.
+ *
+ * Contact: support@cavium.com
+ * Please include "LiquidIO" in the subject.
+ *
+ * Copyright (c) 2003-2016 Cavium, Inc.
+ *
+ * This file is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, Version 2, as
+ * published by the Free Software Foundation.
+ *
+ * This file is distributed in the hope that it will be useful, but
+ * AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
+ * NONINFRINGEMENT. See the GNU General Public License for more details.
+ ***********************************************************************/
+#include <linux/pci.h>
+#include <linux/netdevice.h>
+#include "liquidio_common.h"
+#include "octeon_droq.h"
+#include "octeon_iq.h"
+#include "response_manager.h"
+#include "octeon_device.h"
+#include "octeon_main.h"
+#include "octeon_mailbox.h"
+
+/**
+ * octeon_mbox_read:
+ * @oct: Pointer mailbox
+ *
+ * Reads the 8-bytes of data from the mbox register
+ * Writes back the acknowldgement inidcating completion of read
+ */
+int octeon_mbox_read(struct octeon_mbox *mbox)
+{
+ union octeon_mbox_message msg;
+ int ret = 0;
+
+ spin_lock(&mbox->lock);
+
+ msg.u64 = readq(mbox->mbox_read_reg);
+
+ if ((msg.u64 == OCTEON_PFVFACK) || (msg.u64 == OCTEON_PFVFSIG)) {
+ spin_unlock(&mbox->lock);
+ return 0;
+ }
+
+ if (mbox->state & OCTEON_MBOX_STATE_REQUEST_RECEIVING) {
+ mbox->mbox_req.data[mbox->mbox_req.recv_len - 1] = msg.u64;
+ mbox->mbox_req.recv_len++;
+ } else {
+ if (mbox->state & OCTEON_MBOX_STATE_RESPONSE_RECEIVING) {
+ mbox->mbox_resp.data[mbox->mbox_resp.recv_len - 1] =
+ msg.u64;
+ mbox->mbox_resp.recv_len++;
+ } else {
+ if ((mbox->state & OCTEON_MBOX_STATE_IDLE) &&
+ (msg.s.type == OCTEON_MBOX_REQUEST)) {
+ mbox->state &= ~OCTEON_MBOX_STATE_IDLE;
+ mbox->state |=
+ OCTEON_MBOX_STATE_REQUEST_RECEIVING;
+ mbox->mbox_req.msg.u64 = msg.u64;
+ mbox->mbox_req.q_no = mbox->q_no;
+ mbox->mbox_req.recv_len = 1;
+ } else {
+ if ((mbox->state &
+ OCTEON_MBOX_STATE_RESPONSE_PENDING) &&
+ (msg.s.type == OCTEON_MBOX_RESPONSE)) {
+ mbox->state &=
+ ~OCTEON_MBOX_STATE_RESPONSE_PENDING;
+ mbox->state |=
+ OCTEON_MBOX_STATE_RESPONSE_RECEIVING
+ ;
+ mbox->mbox_resp.msg.u64 = msg.u64;
+ mbox->mbox_resp.q_no = mbox->q_no;
+ mbox->mbox_resp.recv_len = 1;
+ } else {
+ writeq(OCTEON_PFVFERR,
+ mbox->mbox_read_reg);
+ mbox->state |= OCTEON_MBOX_STATE_ERROR;
+ spin_unlock(&mbox->lock);
+ return 1;
+ }
+ }
+ }
+ }
+
+ if (mbox->state & OCTEON_MBOX_STATE_REQUEST_RECEIVING) {
+ if (mbox->mbox_req.recv_len < msg.s.len) {
+ ret = 0;
+ } else {
+ mbox->state &= ~OCTEON_MBOX_STATE_REQUEST_RECEIVING;
+ mbox->state |= OCTEON_MBOX_STATE_REQUEST_RECEIVED;
+ ret = 1;
+ }
+ } else {
+ if (mbox->state & OCTEON_MBOX_STATE_RESPONSE_RECEIVING) {
+ if (mbox->mbox_resp.recv_len < msg.s.len) {
+ ret = 0;
+ } else {
+ mbox->state &=
+ ~OCTEON_MBOX_STATE_RESPONSE_RECEIVING;
+ mbox->state |=
+ OCTEON_MBOX_STATE_RESPONSE_RECEIVED;
+ ret = 1;
+ }
+ } else {
+ WARN_ON(1);
+ }
+ }
+
+ writeq(OCTEON_PFVFACK, mbox->mbox_read_reg);
+
+ spin_unlock(&mbox->lock);
+
+ return ret;
+}
+
+/**
+ * octeon_mbox_write:
+ * @oct: Pointer Octeon Device
+ * @mbox_cmd: Cmd to send to mailbox.
+ *
+ * Populates the queue specific mbox structure
+ * with cmd information.
+ * Write the cmd to mbox register
+ */
+int octeon_mbox_write(struct octeon_device *oct,
+ struct octeon_mbox_cmd *mbox_cmd)
+{
+ struct octeon_mbox *mbox = oct->mbox[mbox_cmd->q_no];
+ u32 count, i, ret = OCTEON_MBOX_STATUS_SUCCESS;
+ unsigned long flags;
+
+ spin_lock_irqsave(&mbox->lock, flags);
+
+ if ((mbox_cmd->msg.s.type == OCTEON_MBOX_RESPONSE) &&
+ !(mbox->state & OCTEON_MBOX_STATE_REQUEST_RECEIVED)) {
+ spin_unlock_irqrestore(&mbox->lock, flags);
+ return OCTEON_MBOX_STATUS_FAILED;
+ }
+
+ if ((mbox_cmd->msg.s.type == OCTEON_MBOX_REQUEST) &&
+ !(mbox->state & OCTEON_MBOX_STATE_IDLE)) {
+ spin_unlock_irqrestore(&mbox->lock, flags);
+ return OCTEON_MBOX_STATUS_BUSY;
+ }
+
+ if (mbox_cmd->msg.s.type == OCTEON_MBOX_REQUEST) {
+ memcpy(&mbox->mbox_resp, mbox_cmd,
+ sizeof(struct octeon_mbox_cmd));
+ mbox->state = OCTEON_MBOX_STATE_RESPONSE_PENDING;
+ }
+
+ spin_unlock_irqrestore(&mbox->lock, flags);
+
+ count = 0;
+
+ while (readq(mbox->mbox_write_reg) != OCTEON_PFVFSIG) {
+ schedule_timeout_uninterruptible(LIO_MBOX_WRITE_WAIT_TIME);
+ if (count++ == LIO_MBOX_WRITE_WAIT_CNT) {
+ ret = OCTEON_MBOX_STATUS_FAILED;
+ break;
+ }
+ }
+
+ if (ret == OCTEON_MBOX_STATUS_SUCCESS) {
+ writeq(mbox_cmd->msg.u64, mbox->mbox_write_reg);
+ for (i = 0; i < (u32)(mbox_cmd->msg.s.len - 1); i++) {
+ count = 0;
+ while (readq(mbox->mbox_write_reg) !=
+ OCTEON_PFVFACK) {
+ schedule_timeout_uninterruptible(10);
+ if (count++ == LIO_MBOX_WRITE_WAIT_CNT) {
+ ret = OCTEON_MBOX_STATUS_FAILED;
+ break;
+ }
+ }
+ writeq(mbox_cmd->data[i], mbox->mbox_write_reg);
+ }
+ }
+
+ spin_lock_irqsave(&mbox->lock, flags);
+ if (mbox_cmd->msg.s.type == OCTEON_MBOX_RESPONSE) {
+ mbox->state = OCTEON_MBOX_STATE_IDLE;
+ writeq(OCTEON_PFVFSIG, mbox->mbox_read_reg);
+ } else {
+ if ((!mbox_cmd->msg.s.resp_needed) ||
+ (ret == OCTEON_MBOX_STATUS_FAILED)) {
+ mbox->state &= ~OCTEON_MBOX_STATE_RESPONSE_PENDING;
+ if (!(mbox->state &
+ (OCTEON_MBOX_STATE_REQUEST_RECEIVING |
+ OCTEON_MBOX_STATE_REQUEST_RECEIVED)))
+ mbox->state = OCTEON_MBOX_STATE_IDLE;
+ }
+ }
+ spin_unlock_irqrestore(&mbox->lock, flags);
+
+ return ret;
+}
+
+/**
+ * octeon_mbox_process_cmd:
+ * @mbox: Pointer mailbox
+ * @mbox_cmd: Pointer to command received
+ *
+ * Process the cmd received in mbox
+ */
+static int octeon_mbox_process_cmd(struct octeon_mbox *mbox,
+ struct octeon_mbox_cmd *mbox_cmd)
+{
+ struct octeon_device *oct = mbox->oct_dev;
+
+ switch (mbox_cmd->msg.s.cmd) {
+ case OCTEON_VF_ACTIVE:
+ dev_dbg(&oct->pci_dev->dev, "got vfactive sending data back\n");
+ mbox_cmd->msg.s.type = OCTEON_MBOX_RESPONSE;
+ mbox_cmd->msg.s.resp_needed = 1;
+ mbox_cmd->msg.s.len = 2;
+ mbox_cmd->data[0] = 0; /* VF version is in mbox_cmd->data[0] */
+ ((struct lio_version *)&mbox_cmd->data[0])->major =
+ LIQUIDIO_BASE_MAJOR_VERSION;
+ ((struct lio_version *)&mbox_cmd->data[0])->minor =
+ LIQUIDIO_BASE_MINOR_VERSION;
+ ((struct lio_version *)&mbox_cmd->data[0])->micro =
+ LIQUIDIO_BASE_MICRO_VERSION;
+ memcpy(mbox_cmd->msg.s.params, (uint8_t *)&oct->pfvf_hsword, 6);
+ /* Sending core cofig info to the corresponding active VF.*/
+ octeon_mbox_write(oct, mbox_cmd);
+ break;
+
+ case OCTEON_VF_FLR_REQUEST:
+ dev_info(&oct->pci_dev->dev,
+ "got a request for FLR from VF that owns DPI ring %u\n",
+ mbox->q_no);
+ pcie_capability_set_word(
+ oct->sriov_info.dpiring_to_vfpcidev_lut[mbox->q_no],
+ PCI_EXP_DEVCTL, PCI_EXP_DEVCTL_BCR_FLR);
+ break;
+
+ case OCTEON_PF_CHANGED_VF_MACADDR:
+ if (OCTEON_CN23XX_VF(oct))
+ octeon_pf_changed_vf_macaddr(oct,
+ mbox_cmd->msg.s.params);
+ break;
+
+ default:
+ break;
+ }
+ return 0;
+}
+
+/**
+ *octeon_mbox_process_message:
+ *
+ * Process the received mbox message.
+ */
+int octeon_mbox_process_message(struct octeon_mbox *mbox)
+{
+ struct octeon_mbox_cmd mbox_cmd;
+ unsigned long flags;
+
+ spin_lock_irqsave(&mbox->lock, flags);
+
+ if (mbox->state & OCTEON_MBOX_STATE_ERROR) {
+ if (mbox->state & (OCTEON_MBOX_STATE_RESPONSE_PENDING |
+ OCTEON_MBOX_STATE_RESPONSE_RECEIVING)) {
+ memcpy(&mbox_cmd, &mbox->mbox_resp,
+ sizeof(struct octeon_mbox_cmd));
+ mbox->state = OCTEON_MBOX_STATE_IDLE;
+ writeq(OCTEON_PFVFSIG, mbox->mbox_read_reg);
+ spin_unlock_irqrestore(&mbox->lock, flags);
+ mbox_cmd.recv_status = 1;
+ if (mbox_cmd.fn)
+ mbox_cmd.fn(mbox->oct_dev, &mbox_cmd,
+ mbox_cmd.fn_arg);
+ return 0;
+ }
+
+ mbox->state = OCTEON_MBOX_STATE_IDLE;
+ writeq(OCTEON_PFVFSIG, mbox->mbox_read_reg);
+ spin_unlock_irqrestore(&mbox->lock, flags);
+ return 0;
+ }
+
+ if (mbox->state & OCTEON_MBOX_STATE_RESPONSE_RECEIVED) {
+ memcpy(&mbox_cmd, &mbox->mbox_resp,
+ sizeof(struct octeon_mbox_cmd));
+ mbox->state = OCTEON_MBOX_STATE_IDLE;
+ writeq(OCTEON_PFVFSIG, mbox->mbox_read_reg);
+ spin_unlock_irqrestore(&mbox->lock, flags);
+ mbox_cmd.recv_status = 0;
+ if (mbox_cmd.fn)
+ mbox_cmd.fn(mbox->oct_dev, &mbox_cmd, mbox_cmd.fn_arg);
+ return 0;
+ }
+
+ if (mbox->state & OCTEON_MBOX_STATE_REQUEST_RECEIVED) {
+ memcpy(&mbox_cmd, &mbox->mbox_req,
+ sizeof(struct octeon_mbox_cmd));
+ if (!mbox_cmd.msg.s.resp_needed) {
+ mbox->state &= ~OCTEON_MBOX_STATE_REQUEST_RECEIVED;
+ if (!(mbox->state &&
+ OCTEON_MBOX_STATE_RESPONSE_PENDING))
+ mbox->state = OCTEON_MBOX_STATE_IDLE;
+ writeq(OCTEON_PFVFSIG, mbox->mbox_read_reg);
+ }
+
+ spin_unlock_irqrestore(&mbox->lock, flags);
+ octeon_mbox_process_cmd(mbox, &mbox_cmd);
+ return 0;
+ }
+
+ WARN_ON(1);
+
+ return 0;
+}
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_mailbox.h b/drivers/net/ethernet/cavium/liquidio/octeon_mailbox.h
new file mode 100644
index 0000000..fe60a3e
--- /dev/null
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_mailbox.h
@@ -0,0 +1,115 @@
+/**********************************************************************
+ * Author: Cavium, Inc.
+ *
+ * Contact: support@cavium.com
+ * Please include "LiquidIO" in the subject.
+ *
+ * Copyright (c) 2003-2016 Cavium, Inc.
+ *
+ * This file is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, Version 2, as
+ * published by the Free Software Foundation.
+ *
+ * This file is distributed in the hope that it will be useful, but
+ * AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
+ * NONINFRINGEMENT. See the GNU General Public License for more details.
+ ***********************************************************************/
+#ifndef __MAILBOX_H__
+#define __MAILBOX_H__
+
+/* Macros for Mail Box Communication */
+
+#define OCTEON_MBOX_DATA_MAX 32
+
+#define OCTEON_VF_ACTIVE 0x1
+#define OCTEON_VF_FLR_REQUEST 0x2
+#define OCTEON_PF_CHANGED_VF_MACADDR 0x4
+
+/*Macro for Read acknowldgement*/
+#define OCTEON_PFVFACK 0xffffffffffffffff
+#define OCTEON_PFVFSIG 0x1122334455667788
+#define OCTEON_PFVFERR 0xDEADDEADDEADDEAD
+
+#define LIO_MBOX_WRITE_WAIT_CNT 1000
+#define LIO_MBOX_WRITE_WAIT_TIME 10
+
+enum octeon_mbox_cmd_status {
+ OCTEON_MBOX_STATUS_SUCCESS = 0,
+ OCTEON_MBOX_STATUS_FAILED = 1,
+ OCTEON_MBOX_STATUS_BUSY = 2
+};
+
+enum octeon_mbox_message_type {
+ OCTEON_MBOX_REQUEST = 0,
+ OCTEON_MBOX_RESPONSE = 1
+};
+
+union octeon_mbox_message {
+ u64 u64;
+ struct {
+ u16 type : 1;
+ u16 resp_needed : 1;
+ u16 cmd : 6;
+ u16 len : 8;
+ u8 params[6];
+ } s;
+};
+
+typedef void (*octeon_mbox_callback_t)(void *, void *, void *);
+
+struct octeon_mbox_cmd {
+ union octeon_mbox_message msg;
+ u64 data[OCTEON_MBOX_DATA_MAX];
+ u32 q_no;
+ u32 recv_len;
+ u32 recv_status;
+ octeon_mbox_callback_t fn;
+ void *fn_arg;
+};
+
+enum octeon_mbox_state {
+ OCTEON_MBOX_STATE_IDLE = 1,
+ OCTEON_MBOX_STATE_REQUEST_RECEIVING = 2,
+ OCTEON_MBOX_STATE_REQUEST_RECEIVED = 4,
+ OCTEON_MBOX_STATE_RESPONSE_PENDING = 8,
+ OCTEON_MBOX_STATE_RESPONSE_RECEIVING = 16,
+ OCTEON_MBOX_STATE_RESPONSE_RECEIVED = 16,
+ OCTEON_MBOX_STATE_ERROR = 32
+};
+
+struct octeon_mbox {
+ /** A spinlock to protect access to this q_mbox. */
+ spinlock_t lock;
+
+ struct octeon_device *oct_dev;
+
+ u32 q_no;
+
+ enum octeon_mbox_state state;
+
+ struct cavium_wk mbox_poll_wk;
+
+ /** SLI_MAC_PF_MBOX_INT for PF, SLI_PKT_MBOX_INT for VF. */
+ void *mbox_int_reg;
+
+ /** SLI_PKT_PF_VF_MBOX_SIG(0) for PF, SLI_PKT_PF_VF_MBOX_SIG(1) for VF.
+ */
+ void *mbox_write_reg;
+
+ /** SLI_PKT_PF_VF_MBOX_SIG(1) for PF, SLI_PKT_PF_VF_MBOX_SIG(0) for VF.
+ */
+ void *mbox_read_reg;
+
+ struct octeon_mbox_cmd mbox_req;
+
+ struct octeon_mbox_cmd mbox_resp;
+
+};
+
+int octeon_mbox_read(struct octeon_mbox *mbox);
+int octeon_mbox_write(struct octeon_device *oct,
+ struct octeon_mbox_cmd *mbox_cmd);
+int octeon_mbox_process_message(struct octeon_mbox *mbox);
+
+#endif
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_main.h b/drivers/net/ethernet/cavium/liquidio/octeon_main.h
index 366298f..3cddf28 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_main.h
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_main.h
@@ -66,7 +66,7 @@ void octeon_update_tx_completion_counters(void *buf, int reqtype,
unsigned int *bytes_compl);
void octeon_report_tx_completion_to_bql(void *txq, unsigned int pkts_compl,
unsigned int bytes_compl);
-
+void octeon_pf_changed_vf_macaddr(struct octeon_device *oct, u8 *mac);
/** Swap 8B blocks */
static inline void octeon_swap_8B_data(u64 *data, u32 blocks)
{
--
1.8.3.1
^ permalink raw reply related
* [PATCH net-next V4 5/9] liquidio CN23XX: VF related operations
From: Raghu Vatsavayi @ 2016-11-03 21:35 UTC (permalink / raw)
To: davem
Cc: netdev, Raghu Vatsavayi, Raghu Vatsavayi, Derek Chickles,
Satanand Burla, Felix Manlunas
In-Reply-To: <1478208918-8330-1-git-send-email-rvatsavayi@caviumnetworks.com>
Adds support for VF related operations like mac address vlan
and link changes.
Signed-off-by: Raghu Vatsavayi <raghu.vatsavayi@caviumnetworks.com>
Signed-off-by: Derek Chickles <derek.chickles@caviumnetworks.com>
Signed-off-by: Satanand Burla <satananda.burla@caviumnetworks.com>
Signed-off-by: Felix Manlunas <felix.manlunas@caviumnetworks.com>
---
.../ethernet/cavium/liquidio/cn23xx_pf_device.c | 22 +++
.../ethernet/cavium/liquidio/cn23xx_pf_device.h | 5 +
drivers/net/ethernet/cavium/liquidio/lio_main.c | 214 +++++++++++++++++++++
.../net/ethernet/cavium/liquidio/liquidio_common.h | 5 +
.../net/ethernet/cavium/liquidio/octeon_device.h | 8 +
5 files changed, 254 insertions(+)
diff --git a/drivers/net/ethernet/cavium/liquidio/cn23xx_pf_device.c b/drivers/net/ethernet/cavium/liquidio/cn23xx_pf_device.c
index 79b8c48..cbf9667 100644
--- a/drivers/net/ethernet/cavium/liquidio/cn23xx_pf_device.c
+++ b/drivers/net/ethernet/cavium/liquidio/cn23xx_pf_device.c
@@ -23,6 +23,7 @@
#include <linux/pci.h>
#include <linux/netdevice.h>
#include <linux/vmalloc.h>
+#include <linux/etherdevice.h>
#include "liquidio_common.h"
#include "octeon_droq.h"
#include "octeon_iq.h"
@@ -1410,3 +1411,24 @@ int cn23xx_fw_loaded(struct octeon_device *oct)
val = octeon_read_csr64(oct, CN23XX_SLI_SCRATCH1);
return (val >> 1) & 1ULL;
}
+
+void cn23xx_tell_vf_its_macaddr_changed(struct octeon_device *oct, int vfidx,
+ u8 *mac)
+{
+ if (oct->sriov_info.vf_drv_loaded_mask & BIT_ULL(vfidx)) {
+ struct octeon_mbox_cmd mbox_cmd;
+
+ mbox_cmd.msg.u64 = 0;
+ mbox_cmd.msg.s.type = OCTEON_MBOX_REQUEST;
+ mbox_cmd.msg.s.resp_needed = 0;
+ mbox_cmd.msg.s.cmd = OCTEON_PF_CHANGED_VF_MACADDR;
+ mbox_cmd.msg.s.len = 1;
+ mbox_cmd.recv_len = 0;
+ mbox_cmd.recv_status = 0;
+ mbox_cmd.fn = NULL;
+ mbox_cmd.fn_arg = 0;
+ ether_addr_copy(mbox_cmd.msg.s.params, mac);
+ mbox_cmd.q_no = vfidx * oct->sriov_info.rings_per_vf;
+ octeon_mbox_write(oct, &mbox_cmd);
+ }
+}
diff --git a/drivers/net/ethernet/cavium/liquidio/cn23xx_pf_device.h b/drivers/net/ethernet/cavium/liquidio/cn23xx_pf_device.h
index 21b5c90..cee346a 100644
--- a/drivers/net/ethernet/cavium/liquidio/cn23xx_pf_device.h
+++ b/drivers/net/ethernet/cavium/liquidio/cn23xx_pf_device.h
@@ -29,6 +29,8 @@
#include "cn23xx_pf_regs.h"
+#define LIO_CMD_WAIT_TM 100
+
/* Register address and configuration for a CN23XX devices.
* If device specific changes need to be made then add a struct to include
* device specific fields as shown in the commented section
@@ -56,4 +58,7 @@ int validate_cn23xx_pf_config_info(struct octeon_device *oct,
void cn23xx_dump_pf_initialized_regs(struct octeon_device *oct);
int cn23xx_fw_loaded(struct octeon_device *oct);
+
+void cn23xx_tell_vf_its_macaddr_changed(struct octeon_device *oct, int vfidx,
+ u8 *mac);
#endif
diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c
index 438465b..b297dae 100644
--- a/drivers/net/ethernet/cavium/liquidio/lio_main.c
+++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c
@@ -3567,6 +3567,151 @@ static void liquidio_del_vxlan_port(struct net_device *netdev,
OCTNET_CMD_VXLAN_PORT_DEL);
}
+static int __liquidio_set_vf_mac(struct net_device *netdev, int vfidx,
+ u8 *mac, bool is_admin_assigned)
+{
+ struct lio *lio = GET_LIO(netdev);
+ struct octeon_device *oct = lio->oct_dev;
+ struct octnic_ctrl_pkt nctrl;
+
+ if (!is_valid_ether_addr(mac))
+ return -EINVAL;
+
+ if (vfidx < 0 || vfidx >= oct->sriov_info.max_vfs)
+ return -EINVAL;
+
+ memset(&nctrl, 0, sizeof(struct octnic_ctrl_pkt));
+
+ nctrl.ncmd.u64 = 0;
+ nctrl.ncmd.s.cmd = OCTNET_CMD_CHANGE_MACADDR;
+ /* vfidx is 0 based, but vf_num (param1) is 1 based */
+ nctrl.ncmd.s.param1 = vfidx + 1;
+ nctrl.ncmd.s.param2 = (is_admin_assigned ? 1 : 0);
+ nctrl.ncmd.s.more = 1;
+ nctrl.iq_no = lio->linfo.txpciq[0].s.q_no;
+ nctrl.cb_fn = 0;
+ nctrl.wait_time = LIO_CMD_WAIT_TM;
+
+ nctrl.udd[0] = 0;
+ /* The MAC Address is presented in network byte order. */
+ ether_addr_copy((u8 *)&nctrl.udd[0] + 2, mac);
+
+ oct->sriov_info.vf_macaddr[vfidx] = nctrl.udd[0];
+
+ octnet_send_nic_ctrl_pkt(oct, &nctrl);
+
+ return 0;
+}
+
+static int liquidio_set_vf_mac(struct net_device *netdev, int vfidx, u8 *mac)
+{
+ struct lio *lio = GET_LIO(netdev);
+ struct octeon_device *oct = lio->oct_dev;
+ int retval;
+
+ retval = __liquidio_set_vf_mac(netdev, vfidx, mac, true);
+ if (!retval)
+ cn23xx_tell_vf_its_macaddr_changed(oct, vfidx, mac);
+
+ return retval;
+}
+
+static int liquidio_set_vf_vlan(struct net_device *netdev, int vfidx,
+ u16 vlan, u8 qos, __be16 vlan_proto)
+{
+ struct lio *lio = GET_LIO(netdev);
+ struct octeon_device *oct = lio->oct_dev;
+ struct octnic_ctrl_pkt nctrl;
+ u16 vlantci;
+
+ if (vfidx < 0 || vfidx >= oct->sriov_info.num_vfs_alloced)
+ return -EINVAL;
+
+ if (vlan_proto != htons(ETH_P_8021Q))
+ return -EPROTONOSUPPORT;
+
+ if (vlan >= VLAN_N_VID || qos > 7)
+ return -EINVAL;
+
+ if (vlan)
+ vlantci = vlan | (u16)qos << VLAN_PRIO_SHIFT;
+ else
+ vlantci = 0;
+
+ if (oct->sriov_info.vf_vlantci[vfidx] == vlantci)
+ return 0;
+
+ memset(&nctrl, 0, sizeof(struct octnic_ctrl_pkt));
+
+ if (vlan)
+ nctrl.ncmd.s.cmd = OCTNET_CMD_ADD_VLAN_FILTER;
+ else
+ nctrl.ncmd.s.cmd = OCTNET_CMD_DEL_VLAN_FILTER;
+
+ nctrl.ncmd.s.param1 = vlantci;
+ nctrl.ncmd.s.param2 =
+ vfidx + 1; /* vfidx is 0 based, but vf_num (param2) is 1 based */
+ nctrl.ncmd.s.more = 0;
+ nctrl.iq_no = lio->linfo.txpciq[0].s.q_no;
+ nctrl.cb_fn = 0;
+ nctrl.wait_time = LIO_CMD_WAIT_TM;
+
+ octnet_send_nic_ctrl_pkt(oct, &nctrl);
+
+ oct->sriov_info.vf_vlantci[vfidx] = vlantci;
+
+ return 0;
+}
+
+static int liquidio_get_vf_config(struct net_device *netdev, int vfidx,
+ struct ifla_vf_info *ivi)
+{
+ struct lio *lio = GET_LIO(netdev);
+ struct octeon_device *oct = lio->oct_dev;
+ u8 *macaddr;
+
+ if (vfidx < 0 || vfidx >= oct->sriov_info.num_vfs_alloced)
+ return -EINVAL;
+
+ ivi->vf = vfidx;
+ macaddr = 2 + (u8 *)&oct->sriov_info.vf_macaddr[vfidx];
+ ether_addr_copy(&ivi->mac[0], macaddr);
+ ivi->vlan = oct->sriov_info.vf_vlantci[vfidx] & VLAN_VID_MASK;
+ ivi->qos = oct->sriov_info.vf_vlantci[vfidx] >> VLAN_PRIO_SHIFT;
+ ivi->linkstate = oct->sriov_info.vf_linkstate[vfidx];
+ return 0;
+}
+
+static int liquidio_set_vf_link_state(struct net_device *netdev, int vfidx,
+ int linkstate)
+{
+ struct lio *lio = GET_LIO(netdev);
+ struct octeon_device *oct = lio->oct_dev;
+ struct octnic_ctrl_pkt nctrl;
+
+ if (vfidx < 0 || vfidx >= oct->sriov_info.num_vfs_alloced)
+ return -EINVAL;
+
+ if (oct->sriov_info.vf_linkstate[vfidx] == linkstate)
+ return 0;
+
+ memset(&nctrl, 0, sizeof(struct octnic_ctrl_pkt));
+ nctrl.ncmd.s.cmd = OCTNET_CMD_SET_VF_LINKSTATE;
+ nctrl.ncmd.s.param1 =
+ vfidx + 1; /* vfidx is 0 based, but vf_num (param1) is 1 based */
+ nctrl.ncmd.s.param2 = linkstate;
+ nctrl.ncmd.s.more = 0;
+ nctrl.iq_no = lio->linfo.txpciq[0].s.q_no;
+ nctrl.cb_fn = 0;
+ nctrl.wait_time = LIO_CMD_WAIT_TM;
+
+ octnet_send_nic_ctrl_pkt(oct, &nctrl);
+
+ oct->sriov_info.vf_linkstate[vfidx] = linkstate;
+
+ return 0;
+}
+
static struct net_device_ops lionetdevops = {
.ndo_open = liquidio_open,
.ndo_stop = liquidio_stop,
@@ -3584,6 +3729,10 @@ static void liquidio_del_vxlan_port(struct net_device *netdev,
.ndo_set_features = liquidio_set_features,
.ndo_udp_tunnel_add = liquidio_add_vxlan_port,
.ndo_udp_tunnel_del = liquidio_del_vxlan_port,
+ .ndo_set_vf_mac = liquidio_set_vf_mac,
+ .ndo_set_vf_vlan = liquidio_set_vf_vlan,
+ .ndo_get_vf_config = liquidio_get_vf_config,
+ .ndo_set_vf_link_state = liquidio_set_vf_link_state,
};
/** \brief Entry point for the liquidio module
@@ -3906,6 +4055,22 @@ static int setup_nic_devices(struct octeon_device *octeon_dev)
"if%d gmx: %d hw_addr: 0x%llx\n", i,
lio->linfo.gmxport, CVM_CAST64(lio->linfo.hw_addr));
+ for (j = 0; j < octeon_dev->sriov_info.max_vfs; j++) {
+ u8 vfmac[ETH_ALEN];
+
+ random_ether_addr(&vfmac[0]);
+ if (__liquidio_set_vf_mac(netdev, j,
+ &vfmac[0], false)) {
+ dev_err(&octeon_dev->pci_dev->dev,
+ "Error setting VF%d MAC address\n",
+ j);
+ goto setup_nic_dev_fail;
+ }
+ dev_info(&octeon_dev->pci_dev->dev,
+ "MAC Address %pM is configured for VF %d\n",
+ &vfmac[0], j);
+ }
+
/* 64-bit swap required on LE machines */
octeon_swap_8B_data(&lio->linfo.hw_addr, 1);
for (j = 0; j < 6; j++)
@@ -4199,6 +4364,52 @@ static void nic_starter(struct work_struct *work)
complete(&handshake[oct->octeon_id].started);
}
+static int
+octeon_recv_vf_drv_notice(struct octeon_recv_info *recv_info, void *buf)
+{
+ struct octeon_device *oct = (struct octeon_device *)buf;
+ struct octeon_recv_pkt *recv_pkt = recv_info->recv_pkt;
+ int i, notice, vf_idx;
+ u64 *data, vf_num;
+
+ notice = recv_pkt->rh.r.ossp;
+ data = (u64 *)get_rbd(recv_pkt->buffer_ptr[0]);
+
+ /* the first 64-bit word of data is the vf_num */
+ vf_num = data[0];
+ octeon_swap_8B_data(&vf_num, 1);
+ vf_idx = (int)vf_num - 1;
+
+ if (notice == VF_DRV_LOADED) {
+ if (!(oct->sriov_info.vf_drv_loaded_mask & BIT_ULL(vf_idx))) {
+ oct->sriov_info.vf_drv_loaded_mask |= BIT_ULL(vf_idx);
+ dev_info(&oct->pci_dev->dev,
+ "driver for VF%d was loaded\n", vf_idx);
+ try_module_get(THIS_MODULE);
+ }
+ } else if (notice == VF_DRV_REMOVED) {
+ if (oct->sriov_info.vf_drv_loaded_mask & BIT_ULL(vf_idx)) {
+ oct->sriov_info.vf_drv_loaded_mask &= ~BIT_ULL(vf_idx);
+ dev_info(&oct->pci_dev->dev,
+ "driver for VF%d was removed\n", vf_idx);
+ module_put(THIS_MODULE);
+ }
+ } else if (notice == VF_DRV_MACADDR_CHANGED) {
+ u8 *b = (u8 *)&data[1];
+
+ oct->sriov_info.vf_macaddr[vf_idx] = data[1];
+ dev_info(&oct->pci_dev->dev,
+ "VF driver changed VF%d's MAC address to %02x:%02x:%02x:%02x:%02x:%02x\n",
+ vf_idx, b[2], b[3], b[4], b[5], b[6], b[7]);
+ }
+
+ for (i = 0; i < recv_pkt->buffer_count; i++)
+ recv_buffer_free(recv_pkt->buffer_ptr[i]);
+ octeon_free_recv_info(recv_info);
+
+ return 0;
+}
+
/**
* \brief Device initialization for each Octeon device that is probed
* @param octeon_dev octeon device
@@ -4257,6 +4468,9 @@ static int octeon_device_init(struct octeon_device *octeon_dev)
octeon_core_drv_init,
octeon_dev);
+ octeon_register_dispatch_fn(octeon_dev, OPCODE_NIC,
+ OPCODE_NIC_VF_DRV_NOTICE,
+ octeon_recv_vf_drv_notice, octeon_dev);
INIT_DELAYED_WORK(&octeon_dev->nic_poll_work.work, nic_starter);
octeon_dev->nic_poll_work.ctxptr = (void *)octeon_dev;
schedule_delayed_work(&octeon_dev->nic_poll_work.work,
diff --git a/drivers/net/ethernet/cavium/liquidio/liquidio_common.h b/drivers/net/ethernet/cavium/liquidio/liquidio_common.h
index caeff9a..edf1282 100644
--- a/drivers/net/ethernet/cavium/liquidio/liquidio_common.h
+++ b/drivers/net/ethernet/cavium/liquidio/liquidio_common.h
@@ -89,6 +89,10 @@ enum octeon_tag_type {
#define OPCODE_NIC_TIMESTAMP 0x07
#define OPCODE_NIC_INTRMOD_CFG 0x08
#define OPCODE_NIC_IF_CFG 0x09
+#define OPCODE_NIC_VF_DRV_NOTICE 0x0A
+#define VF_DRV_LOADED 1
+#define VF_DRV_REMOVED -1
+#define VF_DRV_MACADDR_CHANGED 2
#define CORE_DRV_TEST_SCATTER_OP 0xFFF5
@@ -235,6 +239,7 @@ static inline void add_sg_size(struct octeon_sg_entry *sg_entry,
#define OCTNET_CMD_ID_ACTIVE 0x1a
+#define OCTNET_CMD_SET_VF_LINKSTATE 0x1c
#define OCTNET_CMD_VXLAN_PORT_ADD 0x0
#define OCTNET_CMD_VXLAN_PORT_DEL 0x1
#define OCTNET_CMD_RXCSUM_ENABLE 0x0
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_device.h b/drivers/net/ethernet/cavium/liquidio/octeon_device.h
index bdb10a0..70a0d98 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_device.h
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_device.h
@@ -354,6 +354,14 @@ struct octeon_sriov_info {
/*lookup table that maps DPI ring number to VF pci_dev struct pointer*/
struct pci_dev *dpiring_to_vfpcidev_lut[MAX_POSSIBLE_VFS];
+
+ u64 vf_macaddr[MAX_POSSIBLE_VFS];
+
+ u16 vf_vlantci[MAX_POSSIBLE_VFS];
+
+ int vf_linkstate[MAX_POSSIBLE_VFS];
+
+ u64 vf_drv_loaded_mask;
};
struct octeon_ioq_vector {
--
1.8.3.1
^ permalink raw reply related
* [PATCH net-next V4 4/9] liquidio CN23XX: mailbox interrupt processing
From: Raghu Vatsavayi @ 2016-11-03 21:35 UTC (permalink / raw)
To: davem
Cc: netdev, Raghu Vatsavayi, Raghu Vatsavayi, Derek Chickles,
Satanand Burla, Felix Manlunas
In-Reply-To: <1478208918-8330-1-git-send-email-rvatsavayi@caviumnetworks.com>
Adds support for mailbox interrupt processing of various
commands.
Signed-off-by: Raghu Vatsavayi <raghu.vatsavayi@caviumnetworks.com>
Signed-off-by: Derek Chickles <derek.chickles@caviumnetworks.com>
Signed-off-by: Satanand Burla <satananda.burla@caviumnetworks.com>
Signed-off-by: Felix Manlunas <felix.manlunas@caviumnetworks.com>
---
.../ethernet/cavium/liquidio/cn23xx_pf_device.c | 157 +++++++++++++++++++++
drivers/net/ethernet/cavium/liquidio/lio_main.c | 12 ++
.../net/ethernet/cavium/liquidio/octeon_device.c | 1 +
.../net/ethernet/cavium/liquidio/octeon_device.h | 21 ++-
4 files changed, 184 insertions(+), 7 deletions(-)
diff --git a/drivers/net/ethernet/cavium/liquidio/cn23xx_pf_device.c b/drivers/net/ethernet/cavium/liquidio/cn23xx_pf_device.c
index 4c93eac..79b8c48 100644
--- a/drivers/net/ethernet/cavium/liquidio/cn23xx_pf_device.c
+++ b/drivers/net/ethernet/cavium/liquidio/cn23xx_pf_device.c
@@ -30,6 +30,7 @@
#include "octeon_device.h"
#include "cn23xx_pf_device.h"
#include "octeon_main.h"
+#include "octeon_mailbox.h"
#define RESET_NOTDONE 0
#define RESET_DONE 1
@@ -677,6 +678,118 @@ static void cn23xx_setup_oq_regs(struct octeon_device *oct, u32 oq_no)
}
}
+static void cn23xx_pf_mbox_thread(struct work_struct *work)
+{
+ struct cavium_wk *wk = (struct cavium_wk *)work;
+ struct octeon_mbox *mbox = (struct octeon_mbox *)wk->ctxptr;
+ struct octeon_device *oct = mbox->oct_dev;
+ u64 mbox_int_val, val64;
+ u32 q_no, i;
+
+ if (oct->rev_id < OCTEON_CN23XX_REV_1_1) {
+ /*read and clear by writing 1*/
+ mbox_int_val = readq(mbox->mbox_int_reg);
+ writeq(mbox_int_val, mbox->mbox_int_reg);
+
+ for (i = 0; i < oct->sriov_info.num_vfs_alloced; i++) {
+ q_no = i * oct->sriov_info.rings_per_vf;
+
+ val64 = readq(oct->mbox[q_no]->mbox_write_reg);
+
+ if (val64 && (val64 != OCTEON_PFVFACK)) {
+ if (octeon_mbox_read(oct->mbox[q_no]))
+ octeon_mbox_process_message(
+ oct->mbox[q_no]);
+ }
+ }
+
+ schedule_delayed_work(&wk->work, msecs_to_jiffies(10));
+ } else {
+ octeon_mbox_process_message(mbox);
+ }
+}
+
+static int cn23xx_setup_pf_mbox(struct octeon_device *oct)
+{
+ struct octeon_mbox *mbox = NULL;
+ u16 mac_no = oct->pcie_port;
+ u16 pf_num = oct->pf_num;
+ u32 q_no, i;
+
+ if (!oct->sriov_info.max_vfs)
+ return 0;
+
+ for (i = 0; i < oct->sriov_info.max_vfs; i++) {
+ q_no = i * oct->sriov_info.rings_per_vf;
+
+ mbox = vmalloc(sizeof(*mbox));
+ if (!mbox)
+ goto free_mbox;
+
+ memset(mbox, 0, sizeof(struct octeon_mbox));
+
+ spin_lock_init(&mbox->lock);
+
+ mbox->oct_dev = oct;
+
+ mbox->q_no = q_no;
+
+ mbox->state = OCTEON_MBOX_STATE_IDLE;
+
+ /* PF mbox interrupt reg */
+ mbox->mbox_int_reg = (u8 *)oct->mmio[0].hw_addr +
+ CN23XX_SLI_MAC_PF_MBOX_INT(mac_no, pf_num);
+
+ /* PF writes into SIG0 reg */
+ mbox->mbox_write_reg = (u8 *)oct->mmio[0].hw_addr +
+ CN23XX_SLI_PKT_PF_VF_MBOX_SIG(q_no, 0);
+
+ /* PF reads from SIG1 reg */
+ mbox->mbox_read_reg = (u8 *)oct->mmio[0].hw_addr +
+ CN23XX_SLI_PKT_PF_VF_MBOX_SIG(q_no, 1);
+
+ /*Mail Box Thread creation*/
+ INIT_DELAYED_WORK(&mbox->mbox_poll_wk.work,
+ cn23xx_pf_mbox_thread);
+ mbox->mbox_poll_wk.ctxptr = (void *)mbox;
+
+ oct->mbox[q_no] = mbox;
+
+ writeq(OCTEON_PFVFSIG, mbox->mbox_read_reg);
+ }
+
+ if (oct->rev_id < OCTEON_CN23XX_REV_1_1)
+ schedule_delayed_work(&oct->mbox[0]->mbox_poll_wk.work,
+ msecs_to_jiffies(0));
+
+ return 0;
+
+free_mbox:
+ while (i) {
+ i--;
+ vfree(oct->mbox[i]);
+ }
+
+ return 1;
+}
+
+static int cn23xx_free_pf_mbox(struct octeon_device *oct)
+{
+ u32 q_no, i;
+
+ if (!oct->sriov_info.max_vfs)
+ return 0;
+
+ for (i = 0; i < oct->sriov_info.max_vfs; i++) {
+ q_no = i * oct->sriov_info.rings_per_vf;
+ cancel_delayed_work_sync(
+ &oct->mbox[q_no]->mbox_poll_wk.work);
+ vfree(oct->mbox[q_no]);
+ }
+
+ return 0;
+}
+
static int cn23xx_enable_io_queues(struct octeon_device *oct)
{
u64 reg_val;
@@ -871,6 +984,29 @@ static u64 cn23xx_pf_msix_interrupt_handler(void *dev)
return ret;
}
+static void cn23xx_handle_pf_mbox_intr(struct octeon_device *oct)
+{
+ struct delayed_work *work;
+ u64 mbox_int_val;
+ u32 i, q_no;
+
+ mbox_int_val = readq(oct->mbox[0]->mbox_int_reg);
+
+ for (i = 0; i < oct->sriov_info.num_vfs_alloced; i++) {
+ q_no = i * oct->sriov_info.rings_per_vf;
+
+ if (mbox_int_val & BIT_ULL(q_no)) {
+ writeq(BIT_ULL(q_no),
+ oct->mbox[0]->mbox_int_reg);
+ if (octeon_mbox_read(oct->mbox[q_no])) {
+ work = &oct->mbox[q_no]->mbox_poll_wk.work;
+ schedule_delayed_work(work,
+ msecs_to_jiffies(0));
+ }
+ }
+ }
+}
+
static irqreturn_t cn23xx_interrupt_handler(void *dev)
{
struct octeon_device *oct = (struct octeon_device *)dev;
@@ -886,6 +1022,10 @@ static irqreturn_t cn23xx_interrupt_handler(void *dev)
dev_err(&oct->pci_dev->dev, "OCTEON[%d]: Error Intr: 0x%016llx\n",
oct->octeon_id, CVM_CAST64(intr64));
+ /* When VFs write into MBOX_SIG2 reg,these intr is set in PF */
+ if (intr64 & CN23XX_INTR_VF_MBOX)
+ cn23xx_handle_pf_mbox_intr(oct);
+
if (oct->msix_on != LIO_FLAG_MSIX_ENABLED) {
if (intr64 & CN23XX_INTR_PKT_DATA)
oct->int_status |= OCT_DEV_INTR_PKT_DATA;
@@ -976,6 +1116,13 @@ static void cn23xx_enable_pf_interrupt(struct octeon_device *oct, u8 intr_flag)
intr_val = readq(cn23xx->intr_enb_reg64);
intr_val |= CN23XX_INTR_PKT_DATA;
writeq(intr_val, cn23xx->intr_enb_reg64);
+ } else if ((intr_flag & OCTEON_MBOX_INTR) &&
+ (oct->sriov_info.max_vfs > 0)) {
+ if (oct->rev_id >= OCTEON_CN23XX_REV_1_1) {
+ intr_val = readq(cn23xx->intr_enb_reg64);
+ intr_val |= CN23XX_INTR_VF_MBOX;
+ writeq(intr_val, cn23xx->intr_enb_reg64);
+ }
}
}
@@ -991,6 +1138,13 @@ static void cn23xx_disable_pf_interrupt(struct octeon_device *oct, u8 intr_flag)
intr_val = readq(cn23xx->intr_enb_reg64);
intr_val &= ~CN23XX_INTR_PKT_DATA;
writeq(intr_val, cn23xx->intr_enb_reg64);
+ } else if ((intr_flag & OCTEON_MBOX_INTR) &&
+ (oct->sriov_info.max_vfs > 0)) {
+ if (oct->rev_id >= OCTEON_CN23XX_REV_1_1) {
+ intr_val = readq(cn23xx->intr_enb_reg64);
+ intr_val &= ~CN23XX_INTR_VF_MBOX;
+ writeq(intr_val, cn23xx->intr_enb_reg64);
+ }
}
}
@@ -1137,6 +1291,9 @@ int setup_cn23xx_octeon_pf_device(struct octeon_device *oct)
oct->fn_list.setup_iq_regs = cn23xx_setup_iq_regs;
oct->fn_list.setup_oq_regs = cn23xx_setup_oq_regs;
+ oct->fn_list.setup_mbox = cn23xx_setup_pf_mbox;
+ oct->fn_list.free_mbox = cn23xx_free_pf_mbox;
+
oct->fn_list.process_interrupt_regs = cn23xx_interrupt_handler;
oct->fn_list.msix_interrupt_handler = cn23xx_pf_msix_interrupt_handler;
diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c
index 29f5985..438465b 100644
--- a/drivers/net/ethernet/cavium/liquidio/lio_main.c
+++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c
@@ -1441,6 +1441,12 @@ static void octeon_destroy_resources(struct octeon_device *oct)
if (OCTEON_CN23XX_PF(oct))
octeon_free_ioq_vector(oct);
+
+ /* fallthrough */
+ case OCT_DEV_MBOX_SETUP_DONE:
+ if (OCTEON_CN23XX_PF(oct))
+ oct->fn_list.free_mbox(oct);
+
/* fallthrough */
case OCT_DEV_IN_RESET:
case OCT_DEV_DROQ_INIT_DONE:
@@ -4307,6 +4313,12 @@ static int octeon_device_init(struct octeon_device *octeon_dev)
atomic_set(&octeon_dev->status, OCT_DEV_DROQ_INIT_DONE);
if (OCTEON_CN23XX_PF(octeon_dev)) {
+ if (octeon_dev->fn_list.setup_mbox(octeon_dev)) {
+ dev_err(&octeon_dev->pci_dev->dev, "OCTEON: Mailbox setup failed\n");
+ return 1;
+ }
+ atomic_set(&octeon_dev->status, OCT_DEV_MBOX_SETUP_DONE);
+
if (octeon_allocate_ioq_vector(octeon_dev)) {
dev_err(&octeon_dev->pci_dev->dev, "OCTEON: ioq vector allocation failed\n");
return 1;
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_device.c b/drivers/net/ethernet/cavium/liquidio/octeon_device.c
index 586b688..7f532fd 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_device.c
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_device.c
@@ -767,6 +767,7 @@ struct octeon_device *octeon_allocate_device(u32 pci_id,
ioq_vector->oct_dev = oct;
ioq_vector->iq_index = i;
ioq_vector->droq_index = i;
+ ioq_vector->mbox = oct->mbox[i];
cpu_num = i % num_online_cpus();
cpumask_set_cpu(cpu_num, &ioq_vector->affinity_mask);
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_device.h b/drivers/net/ethernet/cavium/liquidio/octeon_device.h
index 77a6eb7..bdb10a0 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_device.h
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_device.h
@@ -54,6 +54,7 @@ enum octeon_pci_swap_mode {
};
#define OCTEON_OUTPUT_INTR (2)
+#define OCTEON_MBOX_INTR (4)
#define OCTEON_ALL_INTR 0xff
/*--------------- PCI BAR1 index registers -------------*/
@@ -77,13 +78,14 @@ enum octeon_pci_swap_mode {
#define OCT_DEV_SC_BUFF_POOL_INIT_DONE 0x4
#define OCT_DEV_RESP_LIST_INIT_DONE 0x5
#define OCT_DEV_DROQ_INIT_DONE 0x6
-#define OCT_DEV_IO_QUEUES_DONE 0x7
-#define OCT_DEV_CONSOLE_INIT_DONE 0x8
-#define OCT_DEV_HOST_OK 0x9
-#define OCT_DEV_CORE_OK 0xa
-#define OCT_DEV_RUNNING 0xb
-#define OCT_DEV_IN_RESET 0xc
-#define OCT_DEV_STATE_INVALID 0xd
+#define OCT_DEV_MBOX_SETUP_DONE 0x8
+#define OCT_DEV_IO_QUEUES_DONE 0x9
+#define OCT_DEV_CONSOLE_INIT_DONE 0xa
+#define OCT_DEV_HOST_OK 0xb
+#define OCT_DEV_CORE_OK 0xc
+#define OCT_DEV_RUNNING 0xd
+#define OCT_DEV_IN_RESET 0xe
+#define OCT_DEV_STATE_INVALID 0xf
#define OCT_DEV_STATES OCT_DEV_STATE_INVALID
@@ -209,6 +211,10 @@ struct octeon_fn_list {
irqreturn_t (*process_interrupt_regs)(void *);
u64 (*msix_interrupt_handler)(void *);
+
+ int (*setup_mbox)(struct octeon_device *);
+ int (*free_mbox)(struct octeon_device *);
+
int (*soft_reset)(struct octeon_device *);
int (*setup_device_regs)(struct octeon_device *);
void (*bar1_idx_setup)(struct octeon_device *, u64, u32, int);
@@ -355,6 +361,7 @@ struct octeon_ioq_vector {
int iq_index;
int droq_index;
int vector;
+ struct octeon_mbox *mbox;
struct cpumask affinity_mask;
u32 ioq_num;
};
--
1.8.3.1
^ permalink raw reply related
* [PATCH net-next V4 6/9] liquidio CN23XX: device states
From: Raghu Vatsavayi @ 2016-11-03 21:35 UTC (permalink / raw)
To: davem
Cc: netdev, Raghu Vatsavayi, Raghu Vatsavayi, Derek Chickles,
Satanand Burla, Felix Manlunas
In-Reply-To: <1478208918-8330-1-git-send-email-rvatsavayi@caviumnetworks.com>
Cleaned up resource leaks during destroy resources by
introducing more device states.
Signed-off-by: Raghu Vatsavayi <raghu.vatsavayi@caviumnetworks.com>
Signed-off-by: Derek Chickles <derek.chickles@caviumnetworks.com>
Signed-off-by: Satanand Burla <satananda.burla@caviumnetworks.com>
Signed-off-by: Felix Manlunas <felix.manlunas@caviumnetworks.com>
---
drivers/net/ethernet/cavium/liquidio/lio_main.c | 33 ++++++++++++++++------
.../net/ethernet/cavium/liquidio/octeon_device.c | 6 +++-
.../net/ethernet/cavium/liquidio/octeon_device.h | 29 ++++++++++---------
drivers/net/ethernet/cavium/liquidio/octeon_droq.c | 13 +++++----
drivers/net/ethernet/cavium/liquidio/octeon_main.h | 8 ++++--
.../net/ethernet/cavium/liquidio/request_manager.c | 6 +++-
6 files changed, 64 insertions(+), 31 deletions(-)
diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c
index b297dae..10b7e9b 100644
--- a/drivers/net/ethernet/cavium/liquidio/lio_main.c
+++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c
@@ -766,6 +766,7 @@ static void delete_glists(struct lio *lio)
}
kfree((void *)lio->glist);
+ kfree((void *)lio->glist_lock);
}
/**
@@ -1325,6 +1326,7 @@ static int liquidio_watchdog(void *param)
complete(&first_stage);
if (octeon_device_init(oct_dev)) {
+ complete(&hs->init);
liquidio_remove(pdev);
return -ENOMEM;
}
@@ -1349,7 +1351,15 @@ static int liquidio_watchdog(void *param)
oct_dev->watchdog_task = kthread_create(
liquidio_watchdog, oct_dev,
"liowd/%02hhx:%02hhx.%hhx", bus, device, function);
- wake_up_process(oct_dev->watchdog_task);
+ if (!IS_ERR(oct_dev->watchdog_task)) {
+ wake_up_process(oct_dev->watchdog_task);
+ } else {
+ oct_dev->watchdog_task = NULL;
+ dev_err(&oct_dev->pci_dev->dev,
+ "failed to create kernel_thread\n");
+ liquidio_remove(pdev);
+ return -1;
+ }
}
}
@@ -1413,6 +1423,8 @@ static void octeon_destroy_resources(struct octeon_device *oct)
if (lio_wait_for_oq_pkts(oct))
dev_err(&oct->pci_dev->dev, "OQ had pending packets\n");
+ /* fallthrough */
+ case OCT_DEV_INTR_SET_DONE:
/* Disable interrupts */
oct->fn_list.disable_interrupt(oct, OCTEON_ALL_INTR);
@@ -1439,6 +1451,8 @@ static void octeon_destroy_resources(struct octeon_device *oct)
pci_disable_msi(oct->pci_dev);
}
+ /* fallthrough */
+ case OCT_DEV_MSIX_ALLOC_VECTOR_DONE:
if (OCTEON_CN23XX_PF(oct))
octeon_free_ioq_vector(oct);
@@ -1502,10 +1516,13 @@ static void octeon_destroy_resources(struct octeon_device *oct)
octeon_unmap_pci_barx(oct, 1);
/* fallthrough */
- case OCT_DEV_BEGIN_STATE:
+ case OCT_DEV_PCI_ENABLE_DONE:
+ pci_clear_master(oct->pci_dev);
/* Disable the device, releasing the PCI INT */
pci_disable_device(oct->pci_dev);
+ /* fallthrough */
+ case OCT_DEV_BEGIN_STATE:
/* Nothing to be done here either */
break;
} /* end switch (oct->status) */
@@ -1775,6 +1792,7 @@ static int octeon_pci_os_setup(struct octeon_device *oct)
if (dma_set_mask_and_coherent(&oct->pci_dev->dev, DMA_BIT_MASK(64))) {
dev_err(&oct->pci_dev->dev, "Unexpected DMA device capability\n");
+ pci_disable_device(oct->pci_dev);
return 1;
}
@@ -4429,6 +4447,8 @@ static int octeon_device_init(struct octeon_device *octeon_dev)
if (octeon_pci_os_setup(octeon_dev))
return 1;
+ atomic_set(&octeon_dev->status, OCT_DEV_PCI_ENABLE_DONE);
+
/* Identify the Octeon type and map the BAR address space. */
if (octeon_chip_specific_setup(octeon_dev)) {
dev_err(&octeon_dev->pci_dev->dev, "Chip specific setup failed\n");
@@ -4500,9 +4520,6 @@ static int octeon_device_init(struct octeon_device *octeon_dev)
if (octeon_setup_instr_queues(octeon_dev)) {
dev_err(&octeon_dev->pci_dev->dev,
"instruction queue initialization failed\n");
- /* On error, release any previously allocated queues */
- for (j = 0; j < octeon_dev->num_iqs; j++)
- octeon_delete_instr_queue(octeon_dev, j);
return 1;
}
atomic_set(&octeon_dev->status, OCT_DEV_INSTR_QUEUE_INIT_DONE);
@@ -4518,9 +4535,6 @@ static int octeon_device_init(struct octeon_device *octeon_dev)
if (octeon_setup_output_queues(octeon_dev)) {
dev_err(&octeon_dev->pci_dev->dev, "Output queue initialization failed\n");
- /* Release any previously allocated queues */
- for (j = 0; j < octeon_dev->num_oqs; j++)
- octeon_delete_droq(octeon_dev, j);
return 1;
}
@@ -4537,6 +4551,7 @@ static int octeon_device_init(struct octeon_device *octeon_dev)
dev_err(&octeon_dev->pci_dev->dev, "OCTEON: ioq vector allocation failed\n");
return 1;
}
+ atomic_set(&octeon_dev->status, OCT_DEV_MSIX_ALLOC_VECTOR_DONE);
} else {
/* The input and output queue registers were setup earlier (the
@@ -4564,6 +4579,8 @@ static int octeon_device_init(struct octeon_device *octeon_dev)
/* Enable Octeon device interrupts */
octeon_dev->fn_list.enable_interrupt(octeon_dev, OCTEON_ALL_INTR);
+ atomic_set(&octeon_dev->status, OCT_DEV_INTR_SET_DONE);
+
/* Enable the input and output queues for this Octeon device */
ret = octeon_dev->fn_list.enable_io_queues(octeon_dev);
if (ret) {
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_device.c b/drivers/net/ethernet/cavium/liquidio/octeon_device.c
index 7f532fd..295e2ca 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_device.c
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_device.c
@@ -822,6 +822,7 @@ int octeon_setup_instr_queues(struct octeon_device *oct)
if (octeon_init_instr_queue(oct, txpciq, num_descs)) {
/* prevent memory leak */
vfree(oct->instr_queue[0]);
+ oct->instr_queue[0] = NULL;
return 1;
}
@@ -854,8 +855,11 @@ int octeon_setup_output_queues(struct octeon_device *oct)
if (!oct->droq[0])
return 1;
- if (octeon_init_droq(oct, oq_no, num_descs, desc_size, NULL))
+ if (octeon_init_droq(oct, oq_no, num_descs, desc_size, NULL)) {
+ vfree(oct->droq[oq_no]);
+ oct->droq[oq_no] = NULL;
return 1;
+ }
oct->num_oqs++;
return 0;
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_device.h b/drivers/net/ethernet/cavium/liquidio/octeon_device.h
index 70a0d98..6c18baa 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_device.h
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_device.h
@@ -72,20 +72,23 @@ enum octeon_pci_swap_mode {
* as it is initialized.
*/
#define OCT_DEV_BEGIN_STATE 0x0
-#define OCT_DEV_PCI_MAP_DONE 0x1
-#define OCT_DEV_DISPATCH_INIT_DONE 0x2
-#define OCT_DEV_INSTR_QUEUE_INIT_DONE 0x3
-#define OCT_DEV_SC_BUFF_POOL_INIT_DONE 0x4
-#define OCT_DEV_RESP_LIST_INIT_DONE 0x5
-#define OCT_DEV_DROQ_INIT_DONE 0x6
+#define OCT_DEV_PCI_ENABLE_DONE 0x1
+#define OCT_DEV_PCI_MAP_DONE 0x2
+#define OCT_DEV_DISPATCH_INIT_DONE 0x3
+#define OCT_DEV_INSTR_QUEUE_INIT_DONE 0x4
+#define OCT_DEV_SC_BUFF_POOL_INIT_DONE 0x5
+#define OCT_DEV_RESP_LIST_INIT_DONE 0x6
+#define OCT_DEV_DROQ_INIT_DONE 0x7
#define OCT_DEV_MBOX_SETUP_DONE 0x8
-#define OCT_DEV_IO_QUEUES_DONE 0x9
-#define OCT_DEV_CONSOLE_INIT_DONE 0xa
-#define OCT_DEV_HOST_OK 0xb
-#define OCT_DEV_CORE_OK 0xc
-#define OCT_DEV_RUNNING 0xd
-#define OCT_DEV_IN_RESET 0xe
-#define OCT_DEV_STATE_INVALID 0xf
+#define OCT_DEV_MSIX_ALLOC_VECTOR_DONE 0x9
+#define OCT_DEV_INTR_SET_DONE 0xa
+#define OCT_DEV_IO_QUEUES_DONE 0xb
+#define OCT_DEV_CONSOLE_INIT_DONE 0xc
+#define OCT_DEV_HOST_OK 0xd
+#define OCT_DEV_CORE_OK 0xe
+#define OCT_DEV_RUNNING 0xf
+#define OCT_DEV_IN_RESET 0x10
+#define OCT_DEV_STATE_INVALID 0x11
#define OCT_DEV_STATES OCT_DEV_STATE_INVALID
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_droq.c b/drivers/net/ethernet/cavium/liquidio/octeon_droq.c
index f60e532..886987f 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_droq.c
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_droq.c
@@ -988,7 +988,8 @@ int octeon_create_droq(struct octeon_device *oct,
if (!droq)
droq = vmalloc(sizeof(*droq));
if (!droq)
- goto create_droq_fail;
+ return -1;
+
memset(droq, 0, sizeof(struct octeon_droq));
/*Disable the pkt o/p for this Q */
@@ -996,7 +997,11 @@ int octeon_create_droq(struct octeon_device *oct,
oct->droq[q_no] = droq;
/* Initialize the Droq */
- octeon_init_droq(oct, q_no, num_descs, desc_size, app_ctx);
+ if (octeon_init_droq(oct, q_no, num_descs, desc_size, app_ctx)) {
+ vfree(oct->droq[q_no]);
+ oct->droq[q_no] = NULL;
+ return -1;
+ }
oct->num_oqs++;
@@ -1009,8 +1014,4 @@ int octeon_create_droq(struct octeon_device *oct,
* the same time.
*/
return 0;
-
-create_droq_fail:
- octeon_delete_droq(oct, q_no);
- return -ENOMEM;
}
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_main.h b/drivers/net/ethernet/cavium/liquidio/octeon_main.h
index 3cddf28..3afa61e 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_main.h
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_main.h
@@ -116,7 +116,7 @@ static inline int octeon_map_pci_barx(struct octeon_device *oct,
mapped_len = oct->mmio[baridx].len;
if (!mapped_len)
- return 1;
+ goto err_release_region;
if (max_map_len && (mapped_len > max_map_len))
mapped_len = max_map_len;
@@ -132,11 +132,15 @@ static inline int octeon_map_pci_barx(struct octeon_device *oct,
if (!oct->mmio[baridx].hw_addr) {
dev_err(&oct->pci_dev->dev, "error ioremap for bar %d\n",
baridx);
- return 1;
+ goto err_release_region;
}
oct->mmio[baridx].done = 1;
return 0;
+
+err_release_region:
+ pci_release_region(oct->pci_dev, baridx * 2);
+ return 1;
}
static inline void *
diff --git a/drivers/net/ethernet/cavium/liquidio/request_manager.c b/drivers/net/ethernet/cavium/liquidio/request_manager.c
index 90866bb..a73560c 100644
--- a/drivers/net/ethernet/cavium/liquidio/request_manager.c
+++ b/drivers/net/ethernet/cavium/liquidio/request_manager.c
@@ -157,6 +157,8 @@ int octeon_init_instr_queue(struct octeon_device *oct,
WQ_MEM_RECLAIM,
0);
if (!oct->check_db_wq[iq_no].wq) {
+ vfree(iq->request_list);
+ iq->request_list = NULL;
lio_dma_free(oct, q_size, iq->base_addr, iq->base_addr_dma);
dev_err(&oct->pci_dev->dev, "check db wq create failed for iq %d\n",
iq_no);
@@ -749,8 +751,10 @@ int octeon_setup_sc_buffer_pool(struct octeon_device *oct)
lio_dma_alloc(oct,
SOFT_COMMAND_BUFFER_SIZE,
(dma_addr_t *)&dma_addr);
- if (!sc)
+ if (!sc) {
+ octeon_free_sc_buffer_pool(oct);
return 1;
+ }
sc->dma_addr = dma_addr;
sc->size = SOFT_COMMAND_BUFFER_SIZE;
--
1.8.3.1
^ permalink raw reply related
* [PATCH net-next V4 7/9] liquidio CN23XX: code cleanup
From: Raghu Vatsavayi @ 2016-11-03 21:35 UTC (permalink / raw)
To: davem
Cc: netdev, Raghu Vatsavayi, Raghu Vatsavayi, Derek Chickles,
Satanand Burla, Felix Manlunas
In-Reply-To: <1478208918-8330-1-git-send-email-rvatsavayi@caviumnetworks.com>
Cleaned up unnecessary comments and added some minor macros.
Signed-off-by: Raghu Vatsavayi <raghu.vatsavayi@caviumnetworks.com>
Signed-off-by: Derek Chickles <derek.chickles@caviumnetworks.com>
Signed-off-by: Satanand Burla <satananda.burla@caviumnetworks.com>
Signed-off-by: Felix Manlunas <felix.manlunas@caviumnetworks.com>
---
drivers/net/ethernet/cavium/liquidio/cn66xx_device.c | 13 ++++---------
drivers/net/ethernet/cavium/liquidio/cn66xx_device.h | 4 ++--
drivers/net/ethernet/cavium/liquidio/lio_ethtool.c | 14 ++++----------
drivers/net/ethernet/cavium/liquidio/lio_main.c | 16 +++++-----------
drivers/net/ethernet/cavium/liquidio/liquidio_common.h | 2 --
drivers/net/ethernet/cavium/liquidio/octeon_device.c | 8 ++++----
drivers/net/ethernet/cavium/liquidio/octeon_droq.c | 2 +-
drivers/net/ethernet/cavium/liquidio/octeon_droq.h | 1 -
drivers/net/ethernet/cavium/liquidio/octeon_iq.h | 1 -
drivers/net/ethernet/cavium/liquidio/octeon_main.h | 18 ------------------
drivers/net/ethernet/cavium/liquidio/request_manager.c | 7 ++-----
.../net/ethernet/cavium/liquidio/response_manager.c | 6 +-----
.../net/ethernet/cavium/liquidio/response_manager.h | 1 -
13 files changed, 23 insertions(+), 70 deletions(-)
diff --git a/drivers/net/ethernet/cavium/liquidio/cn66xx_device.c b/drivers/net/ethernet/cavium/liquidio/cn66xx_device.c
index e779af8..1ebc225 100644
--- a/drivers/net/ethernet/cavium/liquidio/cn66xx_device.c
+++ b/drivers/net/ethernet/cavium/liquidio/cn66xx_device.c
@@ -275,7 +275,6 @@ void lio_cn6xxx_setup_iq_regs(struct octeon_device *oct, u32 iq_no)
{
struct octeon_instr_queue *iq = oct->instr_queue[iq_no];
- /* Disable Packet-by-Packet mode; No Parse Mode or Skip length */
octeon_write_csr64(oct, CN6XXX_SLI_IQ_PKT_INSTR_HDR64(iq_no), 0);
/* Write the start of the input queue's ring and its size */
@@ -378,7 +377,7 @@ void lio_cn6xxx_disable_io_queues(struct octeon_device *oct)
/* Reset the doorbell register for each Input queue. */
for (i = 0; i < MAX_OCTEON_INSTR_QUEUES(oct); i++) {
- if (!(oct->io_qmask.iq & (1ULL << i)))
+ if (!(oct->io_qmask.iq & BIT_ULL(i)))
continue;
octeon_write_csr(oct, CN6XXX_SLI_IQ_DOORBELL(i), 0xFFFFFFFF);
d32 = octeon_read_csr(oct, CN6XXX_SLI_IQ_DOORBELL(i));
@@ -400,9 +399,8 @@ void lio_cn6xxx_disable_io_queues(struct octeon_device *oct)
;
/* Reset the doorbell register for each Output queue. */
- /* for (i = 0; i < oct->num_oqs; i++) { */
for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES(oct); i++) {
- if (!(oct->io_qmask.oq & (1ULL << i)))
+ if (!(oct->io_qmask.oq & BIT_ULL(i)))
continue;
octeon_write_csr(oct, CN6XXX_SLI_OQ_PKTS_CREDIT(i), 0xFFFFFFFF);
d32 = octeon_read_csr(oct, CN6XXX_SLI_OQ_PKTS_CREDIT(i));
@@ -537,15 +535,14 @@ static int lio_cn6xxx_process_droq_intr_regs(struct octeon_device *oct)
oct->droq_intr = 0;
- /* for (oq_no = 0; oq_no < oct->num_oqs; oq_no++) { */
for (oq_no = 0; oq_no < MAX_OCTEON_OUTPUT_QUEUES(oct); oq_no++) {
- if (!(droq_mask & (1ULL << oq_no)))
+ if (!(droq_mask & BIT_ULL(oq_no)))
continue;
droq = oct->droq[oq_no];
pkt_count = octeon_droq_check_hw_for_pkts(droq);
if (pkt_count) {
- oct->droq_intr |= (1ULL << oq_no);
+ oct->droq_intr |= BIT_ULL(oq_no);
if (droq->ops.poll_mode) {
u32 value;
u32 reg;
@@ -721,8 +718,6 @@ int lio_setup_cn66xx_octeon_device(struct octeon_device *oct)
int lio_validate_cn6xxx_config_info(struct octeon_device *oct,
struct octeon_config *conf6xxx)
{
- /* int total_instrs = 0; */
-
if (CFG_GET_IQ_MAX_Q(conf6xxx) > CN6XXX_MAX_INPUT_QUEUES) {
dev_err(&oct->pci_dev->dev, "%s: Num IQ (%d) exceeds Max (%d)\n",
__func__, CFG_GET_IQ_MAX_Q(conf6xxx),
diff --git a/drivers/net/ethernet/cavium/liquidio/cn66xx_device.h b/drivers/net/ethernet/cavium/liquidio/cn66xx_device.h
index a40a913..32fbbb2 100644
--- a/drivers/net/ethernet/cavium/liquidio/cn66xx_device.h
+++ b/drivers/net/ethernet/cavium/liquidio/cn66xx_device.h
@@ -96,8 +96,8 @@ void lio_cn6xxx_setup_reg_address(struct octeon_device *oct, void *chip,
struct octeon_reg_list *reg_list);
u32 lio_cn6xxx_coprocessor_clock(struct octeon_device *oct);
u32 lio_cn6xxx_get_oq_ticks(struct octeon_device *oct, u32 time_intr_in_us);
-int lio_setup_cn66xx_octeon_device(struct octeon_device *);
+int lio_setup_cn66xx_octeon_device(struct octeon_device *oct);
int lio_validate_cn6xxx_config_info(struct octeon_device *oct,
- struct octeon_config *);
+ struct octeon_config *conf6xxx);
#endif
diff --git a/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c b/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c
index f163e0a..1674522 100644
--- a/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c
+++ b/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c
@@ -757,9 +757,6 @@ static void lio_set_msglevel(struct net_device *netdev, u32 msglvl)
/*sum of oct->instr_queue[iq_no]->stats.tx_dropped */
data[i++] = CVM_CAST64(netstats->tx_dropped);
- /*data[i++] = CVM_CAST64(stats->multicast); */
- /*data[i++] = CVM_CAST64(stats->collisions); */
-
/* firmware tx stats */
/*per_core_stats[cvmx_get_core_num()].link_stats[mdata->from_ifidx].
*fromhost.fw_total_sent
@@ -910,9 +907,8 @@ static void lio_set_msglevel(struct net_device *netdev, u32 msglvl)
/*lio->link_changes*/
data[i++] = CVM_CAST64(lio->link_changes);
- /* TX -- lio_update_stats(lio); */
for (j = 0; j < MAX_OCTEON_INSTR_QUEUES(oct_dev); j++) {
- if (!(oct_dev->io_qmask.iq & (1ULL << j)))
+ if (!(oct_dev->io_qmask.iq & BIT_ULL(j)))
continue;
/*packets to network port*/
/*# of packets tx to network */
@@ -954,9 +950,8 @@ static void lio_set_msglevel(struct net_device *netdev, u32 msglvl)
}
/* RX */
- /* for (j = 0; j < oct_dev->num_oqs; j++) { */
for (j = 0; j < MAX_OCTEON_OUTPUT_QUEUES(oct_dev); j++) {
- if (!(oct_dev->io_qmask.oq & (1ULL << j)))
+ if (!(oct_dev->io_qmask.oq & BIT_ULL(j)))
continue;
/*packets send to TCP/IP network stack */
@@ -1030,7 +1025,7 @@ static void lio_get_strings(struct net_device *netdev, u32 stringset, u8 *data)
num_iq_stats = ARRAY_SIZE(oct_iq_stats_strings);
for (i = 0; i < MAX_OCTEON_INSTR_QUEUES(oct_dev); i++) {
- if (!(oct_dev->io_qmask.iq & (1ULL << i)))
+ if (!(oct_dev->io_qmask.iq & BIT_ULL(i)))
continue;
for (j = 0; j < num_iq_stats; j++) {
sprintf(data, "tx-%d-%s", i,
@@ -1040,9 +1035,8 @@ static void lio_get_strings(struct net_device *netdev, u32 stringset, u8 *data)
}
num_oq_stats = ARRAY_SIZE(oct_droq_stats_strings);
- /* for (i = 0; i < oct_dev->num_oqs; i++) { */
for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES(oct_dev); i++) {
- if (!(oct_dev->io_qmask.oq & (1ULL << i)))
+ if (!(oct_dev->io_qmask.oq & BIT_ULL(i)))
continue;
for (j = 0; j < num_oq_stats; j++) {
sprintf(data, "rx-%d-%s", i,
diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c
index 10b7e9b..a9f19d2 100644
--- a/drivers/net/ethernet/cavium/liquidio/lio_main.c
+++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c
@@ -19,10 +19,8 @@
* This file may also be available under a different license from Cavium.
* Contact Cavium, Inc. for more information
**********************************************************************/
-#include <linux/version.h>
#include <linux/pci.h>
#include <linux/firmware.h>
-#include <linux/ptp_clock_kernel.h>
#include <net/vxlan.h>
#include <linux/kthread.h>
#include "liquidio_common.h"
@@ -199,9 +197,8 @@ static void octeon_droq_bh(unsigned long pdev)
struct octeon_device_priv *oct_priv =
(struct octeon_device_priv *)oct->priv;
- /* for (q_no = 0; q_no < oct->num_oqs; q_no++) { */
for (q_no = 0; q_no < MAX_OCTEON_OUTPUT_QUEUES(oct); q_no++) {
- if (!(oct->io_qmask.oq & (1ULL << q_no)))
+ if (!(oct->io_qmask.oq & BIT_ULL(q_no)))
continue;
reschedule |= octeon_droq_process_packets(oct, oct->droq[q_no],
MAX_PACKET_BUDGET);
@@ -236,7 +233,7 @@ static int lio_wait_for_oq_pkts(struct octeon_device *oct)
pending_pkts = 0;
for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES(oct); i++) {
- if (!(oct->io_qmask.oq & (1ULL << i)))
+ if (!(oct->io_qmask.oq & BIT_ULL(i)))
continue;
pkt_cnt += octeon_droq_check_hw_for_pkts(oct->droq[i]);
}
@@ -318,7 +315,7 @@ static inline void pcierror_quiesce_device(struct octeon_device *oct)
for (i = 0; i < MAX_OCTEON_INSTR_QUEUES(oct); i++) {
struct octeon_instr_queue *iq;
- if (!(oct->io_qmask.iq & (1ULL << i)))
+ if (!(oct->io_qmask.iq & BIT_ULL(i)))
continue;
iq = oct->instr_queue[i];
@@ -384,7 +381,6 @@ static void stop_pci_io(struct octeon_device *oct)
dev_dbg(&oct->pci_dev->dev, "Device state is now %s\n",
lio_get_state_string(&oct->status));
- /* cn63xx_cleanup_aer_uncorrect_error_status(oct->pci_dev); */
/* making it a common function for all OCTEON models */
cleanup_aer_uncorrect_error_status(oct->pci_dev);
}
@@ -937,7 +933,6 @@ static inline void update_link_status(struct net_device *netdev,
if (lio->linfo.link.s.link_up) {
netif_carrier_on(netdev);
- /* start_txq(netdev); */
txqs_wake(netdev);
} else {
netif_carrier_off(netdev);
@@ -1015,7 +1010,7 @@ static void liquidio_schedule_droq_pkt_handlers(struct octeon_device *oct)
if (oct->int_status & OCT_DEV_INTR_PKT_DATA) {
for (oq_no = 0; oq_no < MAX_OCTEON_OUTPUT_QUEUES(oct);
oq_no++) {
- if (!(oct->droq_intr & (1ULL << oq_no)))
+ if (!(oct->droq_intr & BIT_ULL(oq_no)))
continue;
droq = oct->droq[oq_no];
@@ -1464,7 +1459,7 @@ static void octeon_destroy_resources(struct octeon_device *oct)
/* fallthrough */
case OCT_DEV_IN_RESET:
case OCT_DEV_DROQ_INIT_DONE:
- /*atomic_set(&oct->status, OCT_DEV_DROQ_INIT_DONE);*/
+ /* Wait for any pending operations */
mdelay(100);
for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES(oct); i++) {
if (!(oct->io_qmask.oq & BIT_ULL(i)))
@@ -2455,7 +2450,6 @@ static int liquidio_napi_poll(struct napi_struct *napi, int budget)
* Return back if tx_done is false.
*/
update_txq_status(oct, iq_no);
- /*tx_done = (iq->flush_index == iq->octeon_read_index);*/
} else {
dev_err(&oct->pci_dev->dev, "%s: iq (%d) num invalid\n",
__func__, iq_no);
diff --git a/drivers/net/ethernet/cavium/liquidio/liquidio_common.h b/drivers/net/ethernet/cavium/liquidio/liquidio_common.h
index edf1282..1d2eba7 100644
--- a/drivers/net/ethernet/cavium/liquidio/liquidio_common.h
+++ b/drivers/net/ethernet/cavium/liquidio/liquidio_common.h
@@ -68,8 +68,6 @@ enum octeon_tag_type {
*/
#define OPCODE_CORE 0 /* used for generic core operations */
#define OPCODE_NIC 1 /* used for NIC operations */
-#define OPCODE_LAST OPCODE_NIC
-
/* Subcodes are used by host driver/apps to identify the sub-operation
* for the core. They only need to by unique for a given subsystem.
*/
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_device.c b/drivers/net/ethernet/cavium/liquidio/octeon_device.c
index 295e2ca..278c880 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_device.c
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_device.c
@@ -649,12 +649,12 @@ void octeon_free_device_mem(struct octeon_device *oct)
int i;
for (i = 0; i < MAX_OCTEON_OUTPUT_QUEUES(oct); i++) {
- if (oct->io_qmask.oq & (1ULL << i))
+ if (oct->io_qmask.oq & BIT_ULL(i))
vfree(oct->droq[i]);
}
for (i = 0; i < MAX_OCTEON_INSTR_QUEUES(oct); i++) {
- if (oct->io_qmask.iq & (1ULL << i))
+ if (oct->io_qmask.iq & BIT_ULL(i))
vfree(oct->instr_queue[i]);
}
@@ -1148,7 +1148,7 @@ int octeon_get_tx_qsize(struct octeon_device *oct, u32 q_no)
{
if (oct && (q_no < MAX_OCTEON_INSTR_QUEUES(oct)) &&
- (oct->io_qmask.iq & (1ULL << q_no)))
+ (oct->io_qmask.iq & BIT_ULL(q_no)))
return oct->instr_queue[q_no]->max_count;
return -1;
@@ -1157,7 +1157,7 @@ int octeon_get_tx_qsize(struct octeon_device *oct, u32 q_no)
int octeon_get_rx_qsize(struct octeon_device *oct, u32 q_no)
{
if (oct && (q_no < MAX_OCTEON_OUTPUT_QUEUES(oct)) &&
- (oct->io_qmask.oq & (1ULL << q_no)))
+ (oct->io_qmask.oq & BIT_ULL(q_no)))
return oct->droq[q_no]->max_count;
return -1;
}
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_droq.c b/drivers/net/ethernet/cavium/liquidio/octeon_droq.c
index 886987f..8f6fc95 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_droq.c
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_droq.c
@@ -337,7 +337,7 @@ int octeon_init_droq(struct octeon_device *oct,
/* For 56xx Pass1, this function won't be called, so no checks. */
oct->fn_list.setup_oq_regs(oct, q_no);
- oct->io_qmask.oq |= (1ULL << q_no);
+ oct->io_qmask.oq |= BIT_ULL(q_no);
return 0;
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_droq.h b/drivers/net/ethernet/cavium/liquidio/octeon_droq.h
index 5be002d..d591f5f 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_droq.h
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_droq.h
@@ -121,7 +121,6 @@ struct oct_droq_stats {
/** Num of Packets dropped due to receive path failures. */
u64 rx_dropped;
- /** Num of vxlan packets received; */
u64 rx_vxlan;
/** Num of failures of recv_buffer_alloc() */
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_iq.h b/drivers/net/ethernet/cavium/liquidio/octeon_iq.h
index e4d426b..b493e67 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_iq.h
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_iq.h
@@ -69,7 +69,6 @@ struct oct_iq_stats {
u64 tx_vxlan; /* tunnel */
u64 tx_dmamap_fail;
u64 tx_restart;
- /*u64 tx_timeout_count;*/
};
#define OCT_IQ_STATS_SIZE (sizeof(struct oct_iq_stats))
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_main.h b/drivers/net/ethernet/cavium/liquidio/octeon_main.h
index 3afa61e..db86c63 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_main.h
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_main.h
@@ -207,24 +207,6 @@ static inline int octeon_map_pci_barx(struct octeon_device *oct,
return errno;
}
-static inline void
-sleep_atomic_cond(wait_queue_head_t *waitq, atomic_t *pcond)
-{
- wait_queue_t we;
-
- init_waitqueue_entry(&we, current);
- add_wait_queue(waitq, &we);
- while (!atomic_read(pcond)) {
- set_current_state(TASK_INTERRUPTIBLE);
- if (signal_pending(current))
- goto out;
- schedule();
- }
-out:
- set_current_state(TASK_RUNNING);
- remove_wait_queue(waitq, &we);
-}
-
/* Gives up the CPU for a timeout period.
* Check that the condition is not true before we go to sleep for a
* timeout period.
diff --git a/drivers/net/ethernet/cavium/liquidio/request_manager.c b/drivers/net/ethernet/cavium/liquidio/request_manager.c
index a73560c..f6e1616 100644
--- a/drivers/net/ethernet/cavium/liquidio/request_manager.c
+++ b/drivers/net/ethernet/cavium/liquidio/request_manager.c
@@ -145,7 +145,7 @@ int octeon_init_instr_queue(struct octeon_device *oct,
spin_lock_init(&iq->iq_flush_running_lock);
- oct->io_qmask.iq |= (1ULL << iq_no);
+ oct->io_qmask.iq |= BIT_ULL(iq_no);
/* Set the 32B/64B mode for each input queue */
oct->io_qmask.iq64B |= ((conf->instr_type == 64) << iq_no);
@@ -252,9 +252,8 @@ int lio_wait_for_instr_fetch(struct octeon_device *oct)
do {
instr_cnt = 0;
- /*for (i = 0; i < oct->num_iqs; i++) {*/
for (i = 0; i < MAX_OCTEON_INSTR_QUEUES(oct); i++) {
- if (!(oct->io_qmask.iq & (1ULL << i)))
+ if (!(oct->io_qmask.iq & BIT_ULL(i)))
continue;
pending =
atomic_read(&oct->
@@ -579,8 +578,6 @@ static void check_db_timeout(struct work_struct *work)
/* This is only done here to expedite packets being flushed
* for cases where there are no IQ completion interrupts.
*/
- /*if (iq->do_auto_flush)*/
- /* octeon_flush_iq(oct, iq, 2, 0);*/
return st.status;
}
diff --git a/drivers/net/ethernet/cavium/liquidio/response_manager.c b/drivers/net/ethernet/cavium/liquidio/response_manager.c
index be52178d..9e496c1 100644
--- a/drivers/net/ethernet/cavium/liquidio/response_manager.c
+++ b/drivers/net/ethernet/cavium/liquidio/response_manager.c
@@ -81,11 +81,7 @@ int lio_process_ordered_list(struct octeon_device *octeon_dev,
spin_lock_bh(&ordered_sc_list->lock);
if (ordered_sc_list->head.next == &ordered_sc_list->head) {
- /* ordered_sc_list is empty; there is
- * nothing to process
- */
- spin_unlock_bh
- (&ordered_sc_list->lock);
+ spin_unlock_bh(&ordered_sc_list->lock);
return 1;
}
diff --git a/drivers/net/ethernet/cavium/liquidio/response_manager.h b/drivers/net/ethernet/cavium/liquidio/response_manager.h
index 7a48752..d377faf 100644
--- a/drivers/net/ethernet/cavium/liquidio/response_manager.h
+++ b/drivers/net/ethernet/cavium/liquidio/response_manager.h
@@ -85,7 +85,6 @@ enum {
/** A value of 0x00000000 indicates no error i.e. success */
#define DRIVER_ERROR_NONE 0x00000000
-/** (Major number: 0x0000; Minor Number: 0x0001) */
#define DRIVER_ERROR_REQ_PENDING 0x00000001
#define DRIVER_ERROR_REQ_TIMEOUT 0x00000003
#define DRIVER_ERROR_REQ_EINTR 0x00000004
--
1.8.3.1
^ permalink raw reply related
* [PATCH net-next V4 8/9] liquidio CN23XX: copyrights changes and alignment
From: Raghu Vatsavayi @ 2016-11-03 21:35 UTC (permalink / raw)
To: davem
Cc: netdev, Raghu Vatsavayi, Raghu Vatsavayi, Derek Chickles,
Satanand Burla, Felix Manlunas
In-Reply-To: <1478208918-8330-1-git-send-email-rvatsavayi@caviumnetworks.com>
Updated copyrights comments and also changed some other comments
alignments.
Signed-off-by: Raghu Vatsavayi <raghu.vatsavayi@caviumnetworks.com>
Signed-off-by: Derek Chickles <derek.chickles@caviumnetworks.com>
Signed-off-by: Satanand Burla <satananda.burla@caviumnetworks.com>
Signed-off-by: Felix Manlunas <felix.manlunas@caviumnetworks.com>
---
.../ethernet/cavium/liquidio/cn23xx_pf_device.c | 53 ++++++++++------------
.../ethernet/cavium/liquidio/cn23xx_pf_device.h | 39 +++++++---------
.../net/ethernet/cavium/liquidio/cn23xx_pf_regs.h | 39 +++++++---------
.../net/ethernet/cavium/liquidio/cn66xx_device.c | 36 +++++++--------
.../net/ethernet/cavium/liquidio/cn66xx_device.h | 37 +++++++--------
drivers/net/ethernet/cavium/liquidio/cn66xx_regs.h | 37 +++++++--------
.../net/ethernet/cavium/liquidio/cn68xx_device.c | 36 +++++++--------
.../net/ethernet/cavium/liquidio/cn68xx_device.h | 37 +++++++--------
drivers/net/ethernet/cavium/liquidio/cn68xx_regs.h | 37 +++++++--------
drivers/net/ethernet/cavium/liquidio/lio_core.c | 36 +++++++--------
drivers/net/ethernet/cavium/liquidio/lio_ethtool.c | 42 ++++++++---------
drivers/net/ethernet/cavium/liquidio/lio_main.c | 36 +++++++--------
.../net/ethernet/cavium/liquidio/liquidio_common.h | 37 +++++++--------
.../net/ethernet/cavium/liquidio/liquidio_image.h | 36 +++++++--------
.../net/ethernet/cavium/liquidio/octeon_config.h | 37 +++++++--------
.../net/ethernet/cavium/liquidio/octeon_console.c | 43 ++++++++----------
.../net/ethernet/cavium/liquidio/octeon_device.c | 36 +++++++--------
.../net/ethernet/cavium/liquidio/octeon_device.h | 45 ++++++++----------
drivers/net/ethernet/cavium/liquidio/octeon_droq.c | 36 +++++++--------
drivers/net/ethernet/cavium/liquidio/octeon_droq.h | 17 +++----
drivers/net/ethernet/cavium/liquidio/octeon_iq.h | 21 ++++-----
drivers/net/ethernet/cavium/liquidio/octeon_main.h | 19 +++-----
.../net/ethernet/cavium/liquidio/octeon_mem_ops.c | 5 +-
.../net/ethernet/cavium/liquidio/octeon_mem_ops.h | 5 +-
.../net/ethernet/cavium/liquidio/octeon_network.h | 5 +-
drivers/net/ethernet/cavium/liquidio/octeon_nic.c | 5 +-
drivers/net/ethernet/cavium/liquidio/octeon_nic.h | 5 +-
.../net/ethernet/cavium/liquidio/request_manager.c | 5 +-
.../ethernet/cavium/liquidio/response_manager.c | 5 +-
.../ethernet/cavium/liquidio/response_manager.h | 5 +-
30 files changed, 352 insertions(+), 480 deletions(-)
diff --git a/drivers/net/ethernet/cavium/liquidio/cn23xx_pf_device.c b/drivers/net/ethernet/cavium/liquidio/cn23xx_pf_device.c
index cbf9667..2644a84 100644
--- a/drivers/net/ethernet/cavium/liquidio/cn23xx_pf_device.c
+++ b/drivers/net/ethernet/cavium/liquidio/cn23xx_pf_device.c
@@ -1,27 +1,21 @@
/**********************************************************************
-* Author: Cavium, Inc.
-*
-* Contact: support@cavium.com
-* Please include "LiquidIO" in the subject.
-*
-* Copyright (c) 2003-2015 Cavium, Inc.
-*
-* This file is free software; you can redistribute it and/or modify
-* it under the terms of the GNU General Public License, Version 2, as
-* published by the Free Software Foundation.
-*
-* This file is distributed in the hope that it will be useful, but
-* AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
-* of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
-* NONINFRINGEMENT. See the GNU General Public License for more
-* details.
-*
-* This file may also be available under a different license from Cavium.
-* Contact Cavium, Inc. for more information
-**********************************************************************/
-
+ * Author: Cavium, Inc.
+ *
+ * Contact: support@cavium.com
+ * Please include "LiquidIO" in the subject.
+ *
+ * Copyright (c) 2003-2016 Cavium, Inc.
+ *
+ * This file is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, Version 2, as
+ * published by the Free Software Foundation.
+ *
+ * This file is distributed in the hope that it will be useful, but
+ * AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
+ * NONINFRINGEMENT. See the GNU General Public License for more details.
+ ***********************************************************************/
#include <linux/pci.h>
-#include <linux/netdevice.h>
#include <linux/vmalloc.h>
#include <linux/etherdevice.h>
#include "liquidio_common.h"
@@ -421,10 +415,10 @@ static int cn23xx_pf_setup_global_input_regs(struct octeon_device *oct)
return -1;
/** Set the MAC_NUM and PVF_NUM in IQ_PKT_CONTROL reg
- * for all queues.Only PF can set these bits.
- * bits 29:30 indicate the MAC num.
- * bits 32:47 indicate the PVF num.
- */
+ * for all queues.Only PF can set these bits.
+ * bits 29:30 indicate the MAC num.
+ * bits 32:47 indicate the PVF num.
+ */
for (q_no = 0; q_no < ern; q_no++) {
reg_val = oct->pcie_port << CN23XX_PKT_INPUT_CTL_MAC_NUM_POS;
@@ -547,8 +541,8 @@ static void cn23xx_pf_setup_global_output_regs(struct octeon_device *oct)
writeq(0x40, (u8 *)oct->mmio[0].hw_addr + CN23XX_SLI_OQ_WMARK);
/** Disabling setting OQs in reset when ring has no dorebells
- * enabling this will cause of head of line blocking
- */
+ * enabling this will cause of head of line blocking
+ */
/* Do it only for pass1.1. and pass1.2 */
if ((oct->rev_id == OCTEON_CN23XX_REV_1_0) ||
(oct->rev_id == OCTEON_CN23XX_REV_1_1))
@@ -1385,8 +1379,7 @@ void cn23xx_dump_iq_regs(struct octeon_device *oct)
dev_dbg(&oct->pci_dev->dev, "SLI_PKT[%d]_INPUT_CTL [0x%x]: 0x%016llx\n",
q_no, CN23XX_SLI_IQ_PKT_CONTROL64(q_no),
CVM_CAST64(octeon_read_csr64
- (oct,
- CN23XX_SLI_IQ_PKT_CONTROL64(q_no))));
+ (oct, CN23XX_SLI_IQ_PKT_CONTROL64(q_no))));
}
pci_read_config_dword(oct->pci_dev, CN23XX_CONFIG_PCIE_DEVCTL, ®val);
diff --git a/drivers/net/ethernet/cavium/liquidio/cn23xx_pf_device.h b/drivers/net/ethernet/cavium/liquidio/cn23xx_pf_device.h
index cee346a..2fedd91 100644
--- a/drivers/net/ethernet/cavium/liquidio/cn23xx_pf_device.h
+++ b/drivers/net/ethernet/cavium/liquidio/cn23xx_pf_device.h
@@ -1,28 +1,23 @@
/**********************************************************************
-* Author: Cavium, Inc.
-*
-* Contact: support@cavium.com
-* Please include "LiquidIO" in the subject.
-*
-* Copyright (c) 2003-2015 Cavium, Inc.
-*
-* This file is free software; you can redistribute it and/or modify
-* it under the terms of the GNU General Public License, Version 2, as
-* published by the Free Software Foundation.
-*
-* This file is distributed in the hope that it will be useful, but
-* AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
-* of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
-* NONINFRINGEMENT. See the GNU General Public License for more
-* details.
-*
-* This file may also be available under a different license from Cavium.
-* Contact Cavium, Inc. for more information
-**********************************************************************/
-
+ * Author: Cavium, Inc.
+ *
+ * Contact: support@cavium.com
+ * Please include "LiquidIO" in the subject.
+ *
+ * Copyright (c) 2003-2016 Cavium, Inc.
+ *
+ * This file is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, Version 2, as
+ * published by the Free Software Foundation.
+ *
+ * This file is distributed in the hope that it will be useful, but
+ * AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
+ * NONINFRINGEMENT. See the GNU General Public License for more details.
+ ***********************************************************************/
/*! \file cn23xx_device.h
* \brief Host Driver: Routines that perform CN23XX specific operations.
-*/
+ */
#ifndef __CN23XX_PF_DEVICE_H__
#define __CN23XX_PF_DEVICE_H__
diff --git a/drivers/net/ethernet/cavium/liquidio/cn23xx_pf_regs.h b/drivers/net/ethernet/cavium/liquidio/cn23xx_pf_regs.h
index 03d79d9..680a405 100644
--- a/drivers/net/ethernet/cavium/liquidio/cn23xx_pf_regs.h
+++ b/drivers/net/ethernet/cavium/liquidio/cn23xx_pf_regs.h
@@ -1,29 +1,24 @@
/**********************************************************************
-* Author: Cavium, Inc.
-*
-* Contact: support@cavium.com
-* Please include "LiquidIO" in the subject.
-*
-* Copyright (c) 2003-2015 Cavium, Inc.
-*
-* This file is free software; you can redistribute it and/or modify
-* it under the terms of the GNU General Public License, Version 2, as
-* published by the Free Software Foundation.
-*
-* This file is distributed in the hope that it will be useful, but
-* AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
-* of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
-* NONINFRINGEMENT. See the GNU General Public License for more
-* details.
-*
-* This file may also be available under a different license from Cavium.
-* Contact Cavium, Inc. for more information
-**********************************************************************/
-
+ * Author: Cavium, Inc.
+ *
+ * Contact: support@cavium.com
+ * Please include "LiquidIO" in the subject.
+ *
+ * Copyright (c) 2003-2016 Cavium, Inc.
+ *
+ * This file is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, Version 2, as
+ * published by the Free Software Foundation.
+ *
+ * This file is distributed in the hope that it will be useful, but
+ * AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
+ * NONINFRINGEMENT. See the GNU General Public License for more details.
+ ***********************************************************************/
/*! \file cn23xx_regs.h
* \brief Host Driver: Register Address and Register Mask values for
* Octeon CN23XX devices.
-*/
+ */
#ifndef __CN23XX_PF_REGS_H__
#define __CN23XX_PF_REGS_H__
diff --git a/drivers/net/ethernet/cavium/liquidio/cn66xx_device.c b/drivers/net/ethernet/cavium/liquidio/cn66xx_device.c
index 1ebc225..bdec051 100644
--- a/drivers/net/ethernet/cavium/liquidio/cn66xx_device.c
+++ b/drivers/net/ethernet/cavium/liquidio/cn66xx_device.c
@@ -1,24 +1,20 @@
/**********************************************************************
-* Author: Cavium, Inc.
-*
-* Contact: support@cavium.com
-* Please include "LiquidIO" in the subject.
-*
-* Copyright (c) 2003-2015 Cavium, Inc.
-*
-* This file is free software; you can redistribute it and/or modify
-* it under the terms of the GNU General Public License, Version 2, as
-* published by the Free Software Foundation.
-*
-* This file is distributed in the hope that it will be useful, but
-* AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
-* of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
-* NONINFRINGEMENT. See the GNU General Public License for more
-* details.
-*
-* This file may also be available under a different license from Cavium.
-* Contact Cavium, Inc. for more information
-**********************************************************************/
+ * Author: Cavium, Inc.
+ *
+ * Contact: support@cavium.com
+ * Please include "LiquidIO" in the subject.
+ *
+ * Copyright (c) 2003-2016 Cavium, Inc.
+ *
+ * This file is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, Version 2, as
+ * published by the Free Software Foundation.
+ *
+ * This file is distributed in the hope that it will be useful, but
+ * AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
+ * NONINFRINGEMENT. See the GNU General Public License for more details.
+ ***********************************************************************/
#include <linux/pci.h>
#include <linux/netdevice.h>
#include "liquidio_common.h"
diff --git a/drivers/net/ethernet/cavium/liquidio/cn66xx_device.h b/drivers/net/ethernet/cavium/liquidio/cn66xx_device.h
index 32fbbb2..8ed5713 100644
--- a/drivers/net/ethernet/cavium/liquidio/cn66xx_device.h
+++ b/drivers/net/ethernet/cavium/liquidio/cn66xx_device.h
@@ -1,25 +1,20 @@
/**********************************************************************
-* Author: Cavium, Inc.
-*
-* Contact: support@cavium.com
-* Please include "LiquidIO" in the subject.
-*
-* Copyright (c) 2003-2015 Cavium, Inc.
-*
-* This file is free software; you can redistribute it and/or modify
-* it under the terms of the GNU General Public License, Version 2, as
-* published by the Free Software Foundation.
-*
-* This file is distributed in the hope that it will be useful, but
-* AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
-* of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
-* NONINFRINGEMENT. See the GNU General Public License for more
-* details.
-*
-* This file may also be available under a different license from Cavium.
-* Contact Cavium, Inc. for more information
-**********************************************************************/
-
+ * Author: Cavium, Inc.
+ *
+ * Contact: support@cavium.com
+ * Please include "LiquidIO" in the subject.
+ *
+ * Copyright (c) 2003-2016 Cavium, Inc.
+ *
+ * This file is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, Version 2, as
+ * published by the Free Software Foundation.
+ *
+ * This file is distributed in the hope that it will be useful, but
+ * AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
+ * NONINFRINGEMENT. See the GNU General Public License for more details.
+ ***********************************************************************/
/*! \file cn66xx_device.h
* \brief Host Driver: Routines that perform CN66XX specific operations.
*/
diff --git a/drivers/net/ethernet/cavium/liquidio/cn66xx_regs.h b/drivers/net/ethernet/cavium/liquidio/cn66xx_regs.h
index 5e3aff2..23152c0 100644
--- a/drivers/net/ethernet/cavium/liquidio/cn66xx_regs.h
+++ b/drivers/net/ethernet/cavium/liquidio/cn66xx_regs.h
@@ -1,25 +1,20 @@
/**********************************************************************
-* Author: Cavium, Inc.
-*
-* Contact: support@cavium.com
-* Please include "LiquidIO" in the subject.
-*
-* Copyright (c) 2003-2015 Cavium, Inc.
-*
-* This file is free software; you can redistribute it and/or modify
-* it under the terms of the GNU General Public License, Version 2, as
-* published by the Free Software Foundation.
-*
-* This file is distributed in the hope that it will be useful, but
-* AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
-* of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
-* NONINFRINGEMENT. See the GNU General Public License for more
-* details.
-*
-* This file may also be available under a different license from Cavium.
-* Contact Cavium, Inc. for more information
-**********************************************************************/
-
+ * Author: Cavium, Inc.
+ *
+ * Contact: support@cavium.com
+ * Please include "LiquidIO" in the subject.
+ *
+ * Copyright (c) 2003-2016 Cavium, Inc.
+ *
+ * This file is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, Version 2, as
+ * published by the Free Software Foundation.
+ *
+ * This file is distributed in the hope that it will be useful, but
+ * AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
+ * NONINFRINGEMENT. See the GNU General Public License for more details.
+ ***********************************************************************/
/*! \file cn66xx_regs.h
* \brief Host Driver: Register Address and Register Mask values for
* Octeon CN66XX devices.
diff --git a/drivers/net/ethernet/cavium/liquidio/cn68xx_device.c b/drivers/net/ethernet/cavium/liquidio/cn68xx_device.c
index dbf3566..1e40f1a 100644
--- a/drivers/net/ethernet/cavium/liquidio/cn68xx_device.c
+++ b/drivers/net/ethernet/cavium/liquidio/cn68xx_device.c
@@ -1,24 +1,20 @@
/**********************************************************************
-* Author: Cavium, Inc.
-*
-* Contact: support@cavium.com
-* Please include "LiquidIO" in the subject.
-*
-* Copyright (c) 2003-2015 Cavium, Inc.
-*
-* This file is free software; you can redistribute it and/or modify
-* it under the terms of the GNU General Public License, Version 2, as
-* published by the Free Software Foundation.
-*
-* This file is distributed in the hope that it will be useful, but
-* AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
-* of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
-* NONINFRINGEMENT. See the GNU General Public License for more
-* details.
-*
-* This file may also be available under a different license from Cavium.
-* Contact Cavium, Inc. for more information
-**********************************************************************/
+ * Author: Cavium, Inc.
+ *
+ * Contact: support@cavium.com
+ * Please include "LiquidIO" in the subject.
+ *
+ * Copyright (c) 2003-2016 Cavium, Inc.
+ *
+ * This file is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, Version 2, as
+ * published by the Free Software Foundation.
+ *
+ * This file is distributed in the hope that it will be useful, but
+ * AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
+ * NONINFRINGEMENT. See the GNU General Public License for more details.
+ ***********************************************************************/
#include <linux/pci.h>
#include <linux/netdevice.h>
#include "liquidio_common.h"
diff --git a/drivers/net/ethernet/cavium/liquidio/cn68xx_device.h b/drivers/net/ethernet/cavium/liquidio/cn68xx_device.h
index ea7bdcc..66b8d6b 100644
--- a/drivers/net/ethernet/cavium/liquidio/cn68xx_device.h
+++ b/drivers/net/ethernet/cavium/liquidio/cn68xx_device.h
@@ -1,25 +1,20 @@
/**********************************************************************
-* Author: Cavium, Inc.
-*
-* Contact: support@cavium.com
-* Please include "LiquidIO" in the subject.
-*
-* Copyright (c) 2003-2015 Cavium, Inc.
-*
-* This file is free software; you can redistribute it and/or modify
-* it under the terms of the GNU General Public License, Version 2, as
-* published by the Free Software Foundation.
-*
-* This file is distributed in the hope that it will be useful, but
-* AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
-* of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
-* NONINFRINGEMENT. See the GNU General Public License for more
-* details.
-*
-* This file may also be available under a different license from Cavium.
-* Contact Cavium, Inc. for more information
-**********************************************************************/
-
+ * Author: Cavium, Inc.
+ *
+ * Contact: support@cavium.com
+ * Please include "LiquidIO" in the subject.
+ *
+ * Copyright (c) 2003-2016 Cavium, Inc.
+ *
+ * This file is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, Version 2, as
+ * published by the Free Software Foundation.
+ *
+ * This file is distributed in the hope that it will be useful, but
+ * AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
+ * NONINFRINGEMENT. See the GNU General Public License for more details.
+ ***********************************************************************/
/*! \file cn68xx_device.h
* \brief Host Driver: Routines that perform CN68XX specific operations.
*/
diff --git a/drivers/net/ethernet/cavium/liquidio/cn68xx_regs.h b/drivers/net/ethernet/cavium/liquidio/cn68xx_regs.h
index d45a0f4..0b742f0 100644
--- a/drivers/net/ethernet/cavium/liquidio/cn68xx_regs.h
+++ b/drivers/net/ethernet/cavium/liquidio/cn68xx_regs.h
@@ -1,25 +1,20 @@
/**********************************************************************
-* Author: Cavium, Inc.
-*
-* Contact: support@cavium.com
-* Please include "LiquidIO" in the subject.
-*
-* Copyright (c) 2003-2015 Cavium, Inc.
-*
-* This file is free software; you can redistribute it and/or modify
-* it under the terms of the GNU General Public License, Version 2, as
-* published by the Free Software Foundation.
-*
-* This file is distributed in the hope that it will be useful, but
-* AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
-* of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
-* NONINFRINGEMENT. See the GNU General Public License for more
-* details.
-*
-* This file may also be available under a different license from Cavium.
-* Contact Cavium, Inc. for more information
-**********************************************************************/
-
+ * Author: Cavium, Inc.
+ *
+ * Contact: support@cavium.com
+ * Please include "LiquidIO" in the subject.
+ *
+ * Copyright (c) 2003-2016 Cavium, Inc.
+ *
+ * This file is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, Version 2, as
+ * published by the Free Software Foundation.
+ *
+ * This file is distributed in the hope that it will be useful, but
+ * AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
+ * NONINFRINGEMENT. See the GNU General Public License for more details.
+ ***********************************************************************/
/*! \file cn68xx_regs.h
* \brief Host Driver: Register Address and Register Mask values for
* Octeon CN68XX devices. The register map for CN66XX is the same
diff --git a/drivers/net/ethernet/cavium/liquidio/lio_core.c b/drivers/net/ethernet/cavium/liquidio/lio_core.c
index e6026df..b28e740 100644
--- a/drivers/net/ethernet/cavium/liquidio/lio_core.c
+++ b/drivers/net/ethernet/cavium/liquidio/lio_core.c
@@ -1,24 +1,20 @@
/**********************************************************************
-* Author: Cavium, Inc.
-*
-* Contact: support@cavium.com
-* Please include "LiquidIO" in the subject.
-*
-* Copyright (c) 2003-2015 Cavium, Inc.
-*
-* This file is free software; you can redistribute it and/or modify
-* it under the terms of the GNU General Public License, Version 2, as
-* published by the Free Software Foundation.
-*
-* This file is distributed in the hope that it will be useful, but
-* AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
-* of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
-* NONINFRINGEMENT. See the GNU General Public License for more
-* details.
-*
-* This file may also be available under a different license from Cavium.
-* Contact Cavium, Inc. for more information
-**********************************************************************/
+ * Author: Cavium, Inc.
+ *
+ * Contact: support@cavium.com
+ * Please include "LiquidIO" in the subject.
+ *
+ * Copyright (c) 2003-2016 Cavium, Inc.
+ *
+ * This file is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, Version 2, as
+ * published by the Free Software Foundation.
+ *
+ * This file is distributed in the hope that it will be useful, but
+ * AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
+ * NONINFRINGEMENT. See the GNU General Public License for more details.
+ ***********************************************************************/
#include <linux/pci.h>
#include <linux/if_vlan.h>
#include "liquidio_common.h"
diff --git a/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c b/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c
index 1674522..c7c39a0 100644
--- a/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c
+++ b/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c
@@ -1,24 +1,20 @@
/**********************************************************************
-* Author: Cavium, Inc.
-*
-* Contact: support@cavium.com
-* Please include "LiquidIO" in the subject.
-*
-* Copyright (c) 2003-2015 Cavium, Inc.
-*
-* This file is free software; you can redistribute it and/or modify
-* it under the terms of the GNU General Public License, Version 2, as
-* published by the Free Software Foundation.
-*
-* This file is distributed in the hope that it will be useful, but
-* AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
-* of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
-* NONINFRINGEMENT. See the GNU General Public License for more
-* details.
-*
-* This file may also be available under a different license from Cavium.
-* Contact Cavium, Inc. for more information
-**********************************************************************/
+ * Author: Cavium, Inc.
+ *
+ * Contact: support@cavium.com
+ * Please include "LiquidIO" in the subject.
+ *
+ * Copyright (c) 2003-2016 Cavium, Inc.
+ *
+ * This file is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, Version 2, as
+ * published by the Free Software Foundation.
+ *
+ * This file is distributed in the hope that it will be useful, but
+ * AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
+ * NONINFRINGEMENT. See the GNU General Public License for more details.
+ ***********************************************************************/
#include <linux/netdevice.h>
#include <linux/net_tstamp.h>
#include <linux/pci.h>
@@ -87,9 +83,9 @@ enum {
"tx_bytes",
"rx_errors", /*jabber_err+l2_err+frame_err */
"tx_errors", /*fw_err_pko+fw_err_link+fw_err_drop */
- "rx_dropped", /*st->fromwire.total_rcvd - st->fromwire.fw_total_rcvd
- *+st->fromwire.dmac_drop + st->fromwire.fw_err_drop
- */
+ "rx_dropped", /*st->fromwire.total_rcvd - st->fromwire.fw_total_rcvd +
+ *st->fromwire.dmac_drop + st->fromwire.fw_err_drop
+ */
"tx_dropped",
"tx_total_sent",
diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c
index a9f19d2..bf0373a 100644
--- a/drivers/net/ethernet/cavium/liquidio/lio_main.c
+++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c
@@ -1,24 +1,20 @@
/**********************************************************************
-* Author: Cavium, Inc.
-*
-* Contact: support@cavium.com
-* Please include "LiquidIO" in the subject.
-*
-* Copyright (c) 2003-2015 Cavium, Inc.
-*
-* This file is free software; you can redistribute it and/or modify
-* it under the terms of the GNU General Public License, Version 2, as
-* published by the Free Software Foundation.
-*
-* This file is distributed in the hope that it will be useful, but
-* AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
-* of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
-* NONINFRINGEMENT. See the GNU General Public License for more
-* details.
-*
-* This file may also be available under a different license from Cavium.
-* Contact Cavium, Inc. for more information
-**********************************************************************/
+ * Author: Cavium, Inc.
+ *
+ * Contact: support@cavium.com
+ * Please include "LiquidIO" in the subject.
+ *
+ * Copyright (c) 2003-2016 Cavium, Inc.
+ *
+ * This file is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, Version 2, as
+ * published by the Free Software Foundation.
+ *
+ * This file is distributed in the hope that it will be useful, but
+ * AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
+ * NONINFRINGEMENT. See the GNU General Public License for more details.
+ ***********************************************************************/
#include <linux/pci.h>
#include <linux/firmware.h>
#include <net/vxlan.h>
diff --git a/drivers/net/ethernet/cavium/liquidio/liquidio_common.h b/drivers/net/ethernet/cavium/liquidio/liquidio_common.h
index 1d2eba7..82a4eed 100644
--- a/drivers/net/ethernet/cavium/liquidio/liquidio_common.h
+++ b/drivers/net/ethernet/cavium/liquidio/liquidio_common.h
@@ -1,25 +1,20 @@
/**********************************************************************
-* Author: Cavium, Inc.
-*
-* Contact: support@cavium.com
-* Please include "LiquidIO" in the subject.
-*
-* Copyright (c) 2003-2015 Cavium, Inc.
-*
-* This file is free software; you can redistribute it and/or modify
-* it under the terms of the GNU General Public License, Version 2, as
-* published by the Free Software Foundation.
-*
-* This file is distributed in the hope that it will be useful, but
-* AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
-* of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
-* NONINFRINGEMENT. See the GNU General Public License for more
-* details.
-*
-* This file may also be available under a different license from Cavium.
-* Contact Cavium, Inc. for more information
-**********************************************************************/
-
+ * Author: Cavium, Inc.
+ *
+ * Contact: support@cavium.com
+ * Please include "LiquidIO" in the subject.
+ *
+ * Copyright (c) 2003-2016 Cavium, Inc.
+ *
+ * This file is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, Version 2, as
+ * published by the Free Software Foundation.
+ *
+ * This file is distributed in the hope that it will be useful, but
+ * AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
+ * NONINFRINGEMENT. See the GNU General Public License for more details.
+ ***********************************************************************/
/*! \file liquidio_common.h
* \brief Common: Structures and macros used in PCI-NIC package by core and
* host driver.
diff --git a/drivers/net/ethernet/cavium/liquidio/liquidio_image.h b/drivers/net/ethernet/cavium/liquidio/liquidio_image.h
index 93819bd..78a3685 100644
--- a/drivers/net/ethernet/cavium/liquidio/liquidio_image.h
+++ b/drivers/net/ethernet/cavium/liquidio/liquidio_image.h
@@ -1,24 +1,20 @@
/**********************************************************************
-* Author: Cavium, Inc.
-*
-* Contact: support@cavium.com
-* Please include "LiquidIO" in the subject.
-*
-* Copyright (c) 2003-2015 Cavium, Inc.
-*
-* This file is free software; you can redistribute it and/or modify
-* it under the terms of the GNU General Public License, Version 2, as
-* published by the Free Software Foundation.
-*
-* This file is distributed in the hope that it will be useful, but
-* AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
-* of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
-* NONINFRINGEMENT. See the GNU General Public License for more
-* details.
-*
-* This file may also be available under a different license from Cavium.
-* Contact Cavium, Inc. for more information
-**********************************************************************/
+ * Author: Cavium, Inc.
+ *
+ * Contact: support@cavium.com
+ * Please include "LiquidIO" in the subject.
+ *
+ * Copyright (c) 2003-2016 Cavium, Inc.
+ *
+ * This file is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, Version 2, as
+ * published by the Free Software Foundation.
+ *
+ * This file is distributed in the hope that it will be useful, but
+ * AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
+ * NONINFRINGEMENT. See the GNU General Public License for more details.
+ ***********************************************************************/
#ifndef _LIQUIDIO_IMAGE_H_
#define _LIQUIDIO_IMAGE_H_
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_config.h b/drivers/net/ethernet/cavium/liquidio/octeon_config.h
index 71daf82..1cb3514 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_config.h
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_config.h
@@ -1,25 +1,20 @@
/**********************************************************************
-* Author: Cavium, Inc.
-*
-* Contact: support@cavium.com
-* Please include "LiquidIO" in the subject.
-*
-* Copyright (c) 2003-2015 Cavium, Inc.
-*
-* This file is free software; you can redistribute it and/or modify
-* it under the terms of the GNU General Public License, Version 2, as
-* published by the Free Software Foundation.
-*
-* This file is distributed in the hope that it will be useful, but
-* AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
-* of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
-* NONINFRINGEMENT. See the GNU General Public License for more
-* details.
-*
-* This file may also be available under a different license from Cavium.
-* Contact Cavium, Inc. for more information
-**********************************************************************/
-
+ * Author: Cavium, Inc.
+ *
+ * Contact: support@cavium.com
+ * Please include "LiquidIO" in the subject.
+ *
+ * Copyright (c) 2003-2016 Cavium, Inc.
+ *
+ * This file is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, Version 2, as
+ * published by the Free Software Foundation.
+ *
+ * This file is distributed in the hope that it will be useful, but
+ * AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
+ * NONINFRINGEMENT. See the GNU General Public License for more details.
+ ***********************************************************************/
/*! \file octeon_config.h
* \brief Host Driver: Configuration data structures for the host driver.
*/
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_console.c b/drivers/net/ethernet/cavium/liquidio/octeon_console.c
index 01a50f3..c0089ed 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_console.c
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_console.c
@@ -1,25 +1,20 @@
/**********************************************************************
-* Author: Cavium, Inc.
-*
-* Contact: support@cavium.com
-* Please include "LiquidIO" in the subject.
-*
-* Copyright (c) 2003-2015 Cavium, Inc.
-*
-* This file is free software; you can redistribute it and/or modify
-* it under the terms of the GNU General Public License, Version 2, as
-* published by the Free Software Foundation.
-*
-* This file is distributed in the hope that it will be useful, but
-* AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
-* of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
-* NONINFRINGEMENT. See the GNU General Public License for more
-* details.
-*
-* This file may also be available under a different license from Cavium.
-* Contact Cavium, Inc. for more information
-**********************************************************************/
-
+ * Author: Cavium, Inc.
+ *
+ * Contact: support@cavium.com
+ * Please include "LiquidIO" in the subject.
+ *
+ * Copyright (c) 2003-2016 Cavium, Inc.
+ *
+ * This file is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, Version 2, as
+ * published by the Free Software Foundation.
+ *
+ * This file is distributed in the hope that it will be useful, but
+ * AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
+ * NONINFRINGEMENT. See the GNU General Public License for more details.
+ ***********************************************************************/
/**
* @file octeon_console.c
*/
@@ -76,9 +71,9 @@ static int octeon_console_read(struct octeon_device *oct, u32 console_num,
#define OCTEON_CONSOLE_POLL_INTERVAL_MS 100 /* 10 times per second */
/* First three members of cvmx_bootmem_desc are left in original
-** positions for backwards compatibility.
-** Assumes big endian target
-*/
+ * positions for backwards compatibility.
+ * Assumes big endian target
+ */
struct cvmx_bootmem_desc {
/** spinlock to control access to list */
u32 lock;
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_device.c b/drivers/net/ethernet/cavium/liquidio/octeon_device.c
index 278c880..4a30d7c 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_device.c
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_device.c
@@ -1,24 +1,20 @@
/**********************************************************************
-* Author: Cavium, Inc.
-*
-* Contact: support@cavium.com
-* Please include "LiquidIO" in the subject.
-*
-* Copyright (c) 2003-2015 Cavium, Inc.
-*
-* This file is free software; you can redistribute it and/or modify
-* it under the terms of the GNU General Public License, Version 2, as
-* published by the Free Software Foundation.
-*
-* This file is distributed in the hope that it will be useful, but
-* AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
-* of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
-* NONINFRINGEMENT. See the GNU General Public License for more
-* details.
-*
-* This file may also be available under a different license from Cavium.
-* Contact Cavium, Inc. for more information
-**********************************************************************/
+ * Author: Cavium, Inc.
+ *
+ * Contact: support@cavium.com
+ * Please include "LiquidIO" in the subject.
+ *
+ * Copyright (c) 2003-2016 Cavium, Inc.
+ *
+ * This file is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, Version 2, as
+ * published by the Free Software Foundation.
+ *
+ * This file is distributed in the hope that it will be useful, but
+ * AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
+ * NONINFRINGEMENT. See the GNU General Public License for more details.
+ ***********************************************************************/
#include <linux/pci.h>
#include <linux/netdevice.h>
#include <linux/vmalloc.h>
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_device.h b/drivers/net/ethernet/cavium/liquidio/octeon_device.h
index 6c18baa..c59b77e 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_device.h
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_device.h
@@ -1,25 +1,20 @@
/**********************************************************************
-* Author: Cavium, Inc.
-*
-* Contact: support@cavium.com
-* Please include "LiquidIO" in the subject.
-*
-* Copyright (c) 2003-2015 Cavium, Inc.
-*
-* This file is free software; you can redistribute it and/or modify
-* it under the terms of the GNU General Public License, Version 2, as
-* published by the Free Software Foundation.
-*
-* This file is distributed in the hope that it will be useful, but
-* AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
-* of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
-* NONINFRINGEMENT. See the GNU General Public License for more
-* details.
-*
-* This file may also be available under a different license from Cavium.
-* Contact Cavium, Inc. for more information
-**********************************************************************/
-
+ * Author: Cavium, Inc.
+ *
+ * Contact: support@cavium.com
+ * Please include "LiquidIO" in the subject.
+ *
+ * Copyright (c) 2003-2016 Cavium, Inc.
+ *
+ * This file is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, Version 2, as
+ * published by the Free Software Foundation.
+ *
+ * This file is distributed in the hope that it will be useful, but
+ * AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
+ * NONINFRINGEMENT. See the GNU General Public License for more details.
+ ***********************************************************************/
/*! \file octeon_device.h
* \brief Host Driver: This file defines the octeon device structure.
*/
@@ -93,9 +88,9 @@ enum octeon_pci_swap_mode {
#define OCT_DEV_STATES OCT_DEV_STATE_INVALID
/** Octeon Device interrupts
- * These interrupt bits are set in int_status filed of
- * octeon_device structure
- */
+ * These interrupt bits are set in int_status filed of
+ * octeon_device structure
+ */
#define OCT_DEV_INTR_DMA0_FORCE 0x01
#define OCT_DEV_INTR_DMA1_FORCE 0x02
#define OCT_DEV_INTR_PKT_DATA 0x04
@@ -548,7 +543,7 @@ struct oct_intrmod_cmd {
void octeon_init_device_list(int conf_type);
/** Free memory for Input and Output queue structures for a octeon device */
-void octeon_free_device_mem(struct octeon_device *);
+void octeon_free_device_mem(struct octeon_device *oct);
/* Look up a free entry in the octeon_device table and allocate resources
* for the octeon_device structure for an octeon device. Called at init
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_droq.c b/drivers/net/ethernet/cavium/liquidio/octeon_droq.c
index 8f6fc95..d7ba000 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_droq.c
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_droq.c
@@ -1,24 +1,20 @@
/**********************************************************************
-* Author: Cavium, Inc.
-*
-* Contact: support@cavium.com
-* Please include "LiquidIO" in the subject.
-*
-* Copyright (c) 2003-2015 Cavium, Inc.
-*
-* This file is free software; you can redistribute it and/or modify
-* it under the terms of the GNU General Public License, Version 2, as
-* published by the Free Software Foundation.
-*
-* This file is distributed in the hope that it will be useful, but
-* AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
-* of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
-* NONINFRINGEMENT. See the GNU General Public License for more
-* details.
-*
-* This file may also be available under a different license from Cavium.
-* Contact Cavium, Inc. for more information
-**********************************************************************/
+ * Author: Cavium, Inc.
+ *
+ * Contact: support@cavium.com
+ * Please include "LiquidIO" in the subject.
+ *
+ * Copyright (c) 2003-2016 Cavium, Inc.
+ *
+ * This file is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, Version 2, as
+ * published by the Free Software Foundation.
+ *
+ * This file is distributed in the hope that it will be useful, but
+ * AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
+ * NONINFRINGEMENT. See the GNU General Public License for more details.
+ ***********************************************************************/
#include <linux/pci.h>
#include <linux/netdevice.h>
#include <linux/vmalloc.h>
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_droq.h b/drivers/net/ethernet/cavium/liquidio/octeon_droq.h
index d591f5f..e620740 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_droq.h
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_droq.h
@@ -4,7 +4,7 @@
* Contact: support@cavium.com
* Please include "LiquidIO" in the subject.
*
- * Copyright (c) 2003-2015 Cavium, Inc.
+ * Copyright (c) 2003-2016 Cavium, Inc.
*
* This file is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License, Version 2, as
@@ -13,13 +13,8 @@
* This file is distributed in the hope that it will be useful, but
* AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
* of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
- * NONINFRINGEMENT. See the GNU General Public License for more
- * details.
- *
- * This file may also be available under a different license from Cavium.
- * Contact Cavium, Inc. for more information
- **********************************************************************/
-
+ * NONINFRINGEMENT. See the GNU General Public License for more details.
+ ***********************************************************************/
/*! \file octeon_droq.h
* \brief Implementation of Octeon Output queues. "Output" is with
* respect to the Octeon device on the NIC. From this driver's point of
@@ -81,7 +76,7 @@ struct octeon_skb_page_info {
* the Octeon device. Since the descriptor ring keeps physical (bus)
* addresses, this field is required for the driver to keep track of
* the virtual address pointers.
-*/
+ */
struct octeon_recv_buffer {
/** Packet buffer, including metadata. */
void *buffer;
@@ -358,7 +353,7 @@ struct octeon_droq {
* @param q_no - droq no. ranges from 0 - 3.
* @param app_ctx - pointer to application context
* @return Success: 0 Failure: 1
-*/
+ */
int octeon_init_droq(struct octeon_device *oct_dev,
u32 q_no,
u32 num_descs,
@@ -371,7 +366,7 @@ int octeon_init_droq(struct octeon_device *oct_dev,
* @param oct_dev - pointer to the octeon device structure
* @param q_no - droq no. ranges from 0 - 3.
* @return: Success: 0 Failure: 1
-*/
+ */
int octeon_delete_droq(struct octeon_device *oct_dev, u32 q_no);
/** Register a change in droq operations. The ops field has a pointer to a
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_iq.h b/drivers/net/ethernet/cavium/liquidio/octeon_iq.h
index b493e67..c67152a 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_iq.h
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_iq.h
@@ -4,7 +4,7 @@
* Contact: support@cavium.com
* Please include "LiquidIO" in the subject.
*
- * Copyright (c) 2003-2015 Cavium, Inc.
+ * Copyright (c) 2003-2016 Cavium, Inc.
*
* This file is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License, Version 2, as
@@ -13,13 +13,8 @@
* This file is distributed in the hope that it will be useful, but
* AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
* of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
- * NONINFRINGEMENT. See the GNU General Public License for more
- * details.
- *
- * This file may also be available under a different license from Cavium.
- * Contact Cavium, Inc. for more information
- **********************************************************************/
-
+ * NONINFRINGEMENT. See the GNU General Public License for more details.
+ ***********************************************************************/
/*! \file octeon_iq.h
* \brief Host Driver: Implementation of Octeon input queues. "Input" is
* with respect to the Octeon device on the NIC. From this driver's
@@ -77,7 +72,7 @@ struct oct_iq_stats {
* The input queue is used to post raw (instruction) mode data or packet
* data to Octeon device from the host. Each input queue (upto 4) for
* a Octeon device has one such structure to represent it.
-*/
+ */
struct octeon_instr_queue {
struct octeon_device *oct_dev;
@@ -117,8 +112,8 @@ struct octeon_instr_queue {
u32 octeon_read_index;
/** This index aids in finding the window in the queue where Octeon
- * has read the commands.
- */
+ * has read the commands.
+ */
u32 flush_index;
/** This field keeps track of the instructions pending in this queue. */
@@ -149,8 +144,8 @@ struct octeon_instr_queue {
u64 last_db_time;
/** The doorbell timeout. If the doorbell was not rung for this time and
- * fill_cnt is non-zero, ring the doorbell again.
- */
+ * fill_cnt is non-zero, ring the doorbell again.
+ */
u32 db_timeout;
/** Statistics for this input queue. */
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_main.h b/drivers/net/ethernet/cavium/liquidio/octeon_main.h
index db86c63..8cd3891 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_main.h
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_main.h
@@ -4,7 +4,7 @@
* Contact: support@cavium.com
* Please include "LiquidIO" in the subject.
*
- * Copyright (c) 2003-2015 Cavium, Inc.
+ * Copyright (c) 2003-2016 Cavium, Inc.
*
* This file is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License, Version 2, as
@@ -13,13 +13,8 @@
* This file is distributed in the hope that it will be useful, but
* AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
* of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
- * NONINFRINGEMENT. See the GNU General Public License for more
- * details.
- *
- * This file may also be available under a different license from Cavium.
- * Contact Cavium, Inc. for more information
- **********************************************************************/
-
+ * NONINFRINGEMENT. See the GNU General Public License for more details.
+ ***********************************************************************/
/*! \file octeon_main.h
* \brief Host Driver: This file is included by all host driver source files
* to include common definitions.
@@ -78,10 +73,10 @@ static inline void octeon_swap_8B_data(u64 *data, u32 blocks)
}
/**
- * \brief unmaps a PCI BAR
- * @param oct Pointer to Octeon device
- * @param baridx bar index
- */
+ * \brief unmaps a PCI BAR
+ * @param oct Pointer to Octeon device
+ * @param baridx bar index
+ */
static inline void octeon_unmap_pci_barx(struct octeon_device *oct, int baridx)
{
dev_dbg(&oct->pci_dev->dev, "Freeing PCI mapped regions for Bar%d\n",
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_mem_ops.c b/drivers/net/ethernet/cavium/liquidio/octeon_mem_ops.c
index 0dc081a..98d8066d 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_mem_ops.c
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_mem_ops.c
@@ -4,7 +4,7 @@
* Contact: support@cavium.com
* Please include "LiquidIO" in the subject.
*
- * Copyright (c) 2003-2015 Cavium, Inc.
+ * Copyright (c) 2003-2016 Cavium, Inc.
*
* This file is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License, Version 2, as
@@ -15,9 +15,6 @@
* of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
* NONINFRINGEMENT. See the GNU General Public License for more
* details.
- *
- * This file may also be available under a different license from Cavium.
- * Contact Cavium, Inc. for more information
**********************************************************************/
#include <linux/netdevice.h>
#include "liquidio_common.h"
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_mem_ops.h b/drivers/net/ethernet/cavium/liquidio/octeon_mem_ops.h
index 11b1833..bae2fdd 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_mem_ops.h
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_mem_ops.h
@@ -4,7 +4,7 @@
* Contact: support@cavium.com
* Please include "LiquidIO" in the subject.
*
- * Copyright (c) 2003-2015 Cavium, Inc.
+ * Copyright (c) 2003-2016 Cavium, Inc.
*
* This file is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License, Version 2, as
@@ -15,9 +15,6 @@
* of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
* NONINFRINGEMENT. See the GNU General Public License for more
* details.
- *
- * This file may also be available under a different license from Cavium.
- * Contact Cavium, Inc. for more information
**********************************************************************/
/*! \file octeon_mem_ops.h
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_network.h b/drivers/net/ethernet/cavium/liquidio/octeon_network.h
index 54b9665..172b70e 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_network.h
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_network.h
@@ -4,7 +4,7 @@
* Contact: support@cavium.com
* Please include "LiquidIO" in the subject.
*
- * Copyright (c) 2003-2015 Cavium, Inc.
+ * Copyright (c) 2003-2016 Cavium, Inc.
*
* This file is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License, Version 2, as
@@ -15,9 +15,6 @@
* of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
* NONINFRINGEMENT. See the GNU General Public License for more
* details.
- *
- * This file may also be available under a different license from Cavium.
- * Contact Cavium, Inc. for more information
**********************************************************************/
/*! \file octeon_network.h
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_nic.c b/drivers/net/ethernet/cavium/liquidio/octeon_nic.c
index 40ac1fe..c3d6a82 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_nic.c
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_nic.c
@@ -4,7 +4,7 @@
* Contact: support@cavium.com
* Please include "LiquidIO" in the subject.
*
- * Copyright (c) 2003-2015 Cavium, Inc.
+ * Copyright (c) 2003-2016 Cavium, Inc.
*
* This file is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License, Version 2, as
@@ -15,9 +15,6 @@
* of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
* NONINFRINGEMENT. See the GNU General Public License for more
* details.
- *
- * This file may also be available under a different license from Cavium.
- * Contact Cavium, Inc. for more information
**********************************************************************/
#include <linux/pci.h>
#include <linux/netdevice.h>
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_nic.h b/drivers/net/ethernet/cavium/liquidio/octeon_nic.h
index 4b8da67b..b059979 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_nic.h
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_nic.h
@@ -4,7 +4,7 @@
* Contact: support@cavium.com
* Please include "LiquidIO" in the subject.
*
- * Copyright (c) 2003-2015 Cavium, Inc.
+ * Copyright (c) 2003-2016 Cavium, Inc.
*
* This file is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License, Version 2, as
@@ -15,9 +15,6 @@
* of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
* NONINFRINGEMENT. See the GNU General Public License for more
* details.
- *
- * This file may also be available under a different license from Cavium.
- * Contact Cavium, Inc. for more information
**********************************************************************/
/*! \file octeon_nic.h
diff --git a/drivers/net/ethernet/cavium/liquidio/request_manager.c b/drivers/net/ethernet/cavium/liquidio/request_manager.c
index f6e1616..0cc1583 100644
--- a/drivers/net/ethernet/cavium/liquidio/request_manager.c
+++ b/drivers/net/ethernet/cavium/liquidio/request_manager.c
@@ -4,7 +4,7 @@
* Contact: support@cavium.com
* Please include "LiquidIO" in the subject.
*
- * Copyright (c) 2003-2015 Cavium, Inc.
+ * Copyright (c) 2003-2016 Cavium, Inc.
*
* This file is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License, Version 2, as
@@ -15,9 +15,6 @@
* of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
* NONINFRINGEMENT. See the GNU General Public License for more
* details.
- *
- * This file may also be available under a different license from Cavium.
- * Contact Cavium, Inc. for more information
**********************************************************************/
#include <linux/pci.h>
#include <linux/netdevice.h>
diff --git a/drivers/net/ethernet/cavium/liquidio/response_manager.c b/drivers/net/ethernet/cavium/liquidio/response_manager.c
index 9e496c1..fdaf742 100644
--- a/drivers/net/ethernet/cavium/liquidio/response_manager.c
+++ b/drivers/net/ethernet/cavium/liquidio/response_manager.c
@@ -4,7 +4,7 @@
* Contact: support@cavium.com
* Please include "LiquidIO" in the subject.
*
- * Copyright (c) 2003-2015 Cavium, Inc.
+ * Copyright (c) 2003-2016 Cavium, Inc.
*
* This file is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License, Version 2, as
@@ -15,9 +15,6 @@
* of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
* NONINFRINGEMENT. See the GNU General Public License for more
* details.
- *
- * This file may also be available under a different license from Cavium.
- * Contact Cavium, Inc. for more information
**********************************************************************/
#include <linux/pci.h>
#include <linux/netdevice.h>
diff --git a/drivers/net/ethernet/cavium/liquidio/response_manager.h b/drivers/net/ethernet/cavium/liquidio/response_manager.h
index d377faf..cbb2d84 100644
--- a/drivers/net/ethernet/cavium/liquidio/response_manager.h
+++ b/drivers/net/ethernet/cavium/liquidio/response_manager.h
@@ -4,7 +4,7 @@
* Contact: support@cavium.com
* Please include "LiquidIO" in the subject.
*
- * Copyright (c) 2003-2015 Cavium, Inc.
+ * Copyright (c) 2003-2016 Cavium, Inc.
*
* This file is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License, Version 2, as
@@ -15,9 +15,6 @@
* of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
* NONINFRINGEMENT. See the GNU General Public License for more
* details.
- *
- * This file may also be available under a different license from Cavium.
- * Contact Cavium, Inc. for more information
**********************************************************************/
/*! \file response_manager.h
--
1.8.3.1
^ permalink raw reply related
* [PATCH net-next V4 9/9] liquidio CN23XX: fix for new check patch errors
From: Raghu Vatsavayi @ 2016-11-03 21:35 UTC (permalink / raw)
To: davem
Cc: netdev, Raghu Vatsavayi, Raghu Vatsavayi, Derek Chickles,
Satanand Burla, Felix Manlunas
In-Reply-To: <1478208918-8330-1-git-send-email-rvatsavayi@caviumnetworks.com>
New checkpatch script shows some errors with pre-existing
driver. This patch provides fix for those errors.
Signed-off-by: Raghu Vatsavayi <raghu.vatsavayi@caviumnetworks.com>
Signed-off-by: Derek Chickles <derek.chickles@caviumnetworks.com>
Signed-off-by: Satanand Burla <satananda.burla@caviumnetworks.com>
Signed-off-by: Felix Manlunas <felix.manlunas@caviumnetworks.com>
---
.../net/ethernet/cavium/liquidio/cn23xx_pf_regs.h | 12 +--
drivers/net/ethernet/cavium/liquidio/cn66xx_regs.h | 12 +--
.../net/ethernet/cavium/liquidio/cn68xx_device.c | 2 +-
drivers/net/ethernet/cavium/liquidio/lio_ethtool.c | 9 +-
drivers/net/ethernet/cavium/liquidio/lio_main.c | 15 +--
.../net/ethernet/cavium/liquidio/liquidio_common.h | 50 ++++-----
.../net/ethernet/cavium/liquidio/octeon_console.c | 113 ++++++++++-----------
.../net/ethernet/cavium/liquidio/octeon_device.c | 28 ++---
.../net/ethernet/cavium/liquidio/octeon_device.h | 25 +++--
drivers/net/ethernet/cavium/liquidio/octeon_droq.c | 40 ++++----
drivers/net/ethernet/cavium/liquidio/octeon_iq.h | 3 +
.../net/ethernet/cavium/liquidio/octeon_mem_ops.c | 2 +-
.../net/ethernet/cavium/liquidio/octeon_network.h | 6 +-
drivers/net/ethernet/cavium/liquidio/octeon_nic.h | 2 +-
.../net/ethernet/cavium/liquidio/request_manager.c | 16 ++-
15 files changed, 155 insertions(+), 180 deletions(-)
diff --git a/drivers/net/ethernet/cavium/liquidio/cn23xx_pf_regs.h b/drivers/net/ethernet/cavium/liquidio/cn23xx_pf_regs.h
index 680a405..e6d4ad9 100644
--- a/drivers/net/ethernet/cavium/liquidio/cn23xx_pf_regs.h
+++ b/drivers/net/ethernet/cavium/liquidio/cn23xx_pf_regs.h
@@ -58,7 +58,7 @@
#define CN23XX_CONFIG_SRIOV_BAR_START 0x19C
#define CN23XX_CONFIG_SRIOV_BARX(i) \
- (CN23XX_CONFIG_SRIOV_BAR_START + (i * 4))
+ (CN23XX_CONFIG_SRIOV_BAR_START + ((i) * 4))
#define CN23XX_CONFIG_SRIOV_BAR_PF 0x08
#define CN23XX_CONFIG_SRIOV_BAR_64BIT 0x04
#define CN23XX_CONFIG_SRIOV_BAR_IO 0x01
@@ -508,7 +508,7 @@
/* 4 Registers (64 - bit) */
#define CN23XX_SLI_S2M_PORT_CTL_START 0x23D80
#define CN23XX_SLI_S2M_PORTX_CTL(port) \
- (CN23XX_SLI_S2M_PORT_CTL_START + (port * 0x10))
+ (CN23XX_SLI_S2M_PORT_CTL_START + ((port) * 0x10))
#define CN23XX_SLI_MAC_NUMBER 0x20050
@@ -549,26 +549,26 @@
* Provides DMA Engine Queue Enable
*/
#define CN23XX_DPI_DMA_ENG0_ENB 0x0001df0000000080ULL
-#define CN23XX_DPI_DMA_ENG_ENB(eng) (CN23XX_DPI_DMA_ENG0_ENB + (eng * 8))
+#define CN23XX_DPI_DMA_ENG_ENB(eng) (CN23XX_DPI_DMA_ENG0_ENB + ((eng) * 8))
/* 8 register (64-bit) - DPI_DMA(0..7)_REQQ_CTL
* Provides control bits for transaction on 8 Queues
*/
#define CN23XX_DPI_DMA_REQQ0_CTL 0x0001df0000000180ULL
#define CN23XX_DPI_DMA_REQQ_CTL(q_no) \
- (CN23XX_DPI_DMA_REQQ0_CTL + (q_no * 8))
+ (CN23XX_DPI_DMA_REQQ0_CTL + ((q_no) * 8))
/* 6 register (64-bit) - DPI_ENG(0..5)_BUF
* Provides DMA Engine FIFO (Queue) Size
*/
#define CN23XX_DPI_DMA_ENG0_BUF 0x0001df0000000880ULL
#define CN23XX_DPI_DMA_ENG_BUF(eng) \
- (CN23XX_DPI_DMA_ENG0_BUF + (eng * 8))
+ (CN23XX_DPI_DMA_ENG0_BUF + ((eng) * 8))
/* 4 Registers (64-bit) */
#define CN23XX_DPI_SLI_PRT_CFG_START 0x0001df0000000900ULL
#define CN23XX_DPI_SLI_PRTX_CFG(port) \
- (CN23XX_DPI_SLI_PRT_CFG_START + (port * 0x8))
+ (CN23XX_DPI_SLI_PRT_CFG_START + ((port) * 0x8))
/* Masks for DPI_DMA_CONTROL Register */
#define CN23XX_DPI_DMA_COMMIT_MODE BIT_ULL(58)
diff --git a/drivers/net/ethernet/cavium/liquidio/cn66xx_regs.h b/drivers/net/ethernet/cavium/liquidio/cn66xx_regs.h
index 23152c0..b248966 100644
--- a/drivers/net/ethernet/cavium/liquidio/cn66xx_regs.h
+++ b/drivers/net/ethernet/cavium/liquidio/cn66xx_regs.h
@@ -438,10 +438,10 @@
#define CN6XXX_SLI_S2M_PORT0_CTL 0x3D80
#define CN6XXX_SLI_S2M_PORT1_CTL 0x3D90
#define CN6XXX_SLI_S2M_PORTX_CTL(port) \
- (CN6XXX_SLI_S2M_PORT0_CTL + (port * 0x10))
+ (CN6XXX_SLI_S2M_PORT0_CTL + ((port) * 0x10))
#define CN6XXX_SLI_INT_ENB64(port) \
- (CN6XXX_SLI_INT_ENB64_PORT0 + (port * 0x10))
+ (CN6XXX_SLI_INT_ENB64_PORT0 + ((port) * 0x10))
#define CN6XXX_SLI_MAC_NUMBER 0x3E00
@@ -453,7 +453,7 @@
#define CN6XXX_PCI_BAR1_OFFSET 0x8
#define CN6XXX_BAR1_REG(idx, port) \
- (CN6XXX_BAR1_INDEX_START + (port * CN6XXX_PEM_OFFSET) + \
+ (CN6XXX_BAR1_INDEX_START + ((port) * CN6XXX_PEM_OFFSET) + \
(CN6XXX_PCI_BAR1_OFFSET * (idx)))
/*############################ DPI #########################*/
@@ -471,17 +471,17 @@
#define CN6XXX_DPI_DMA_ENG0_ENB 0x0001df0000000080ULL
#define CN6XXX_DPI_DMA_ENG_ENB(q_no) \
- (CN6XXX_DPI_DMA_ENG0_ENB + (q_no * 8))
+ (CN6XXX_DPI_DMA_ENG0_ENB + ((q_no) * 8))
#define CN6XXX_DPI_DMA_ENG0_BUF 0x0001df0000000880ULL
#define CN6XXX_DPI_DMA_ENG_BUF(q_no) \
- (CN6XXX_DPI_DMA_ENG0_BUF + (q_no * 8))
+ (CN6XXX_DPI_DMA_ENG0_BUF + ((q_no) * 8))
#define CN6XXX_DPI_SLI_PRT0_CFG 0x0001df0000000900ULL
#define CN6XXX_DPI_SLI_PRT1_CFG 0x0001df0000000908ULL
#define CN6XXX_DPI_SLI_PRTX_CFG(port) \
- (CN6XXX_DPI_SLI_PRT0_CFG + (port * 0x10))
+ (CN6XXX_DPI_SLI_PRT0_CFG + ((port) * 0x10))
#define CN6XXX_DPI_DMA_COMMIT_MODE BIT_ULL(58)
#define CN6XXX_DPI_DMA_PKT_HP BIT_ULL(57)
diff --git a/drivers/net/ethernet/cavium/liquidio/cn68xx_device.c b/drivers/net/ethernet/cavium/liquidio/cn68xx_device.c
index 1e40f1a..50b533f 100644
--- a/drivers/net/ethernet/cavium/liquidio/cn68xx_device.c
+++ b/drivers/net/ethernet/cavium/liquidio/cn68xx_device.c
@@ -72,7 +72,7 @@ static void lio_cn68xx_setup_pkt_ctl_regs(struct octeon_device *oct)
pktctl = octeon_read_csr64(oct, CN6XXX_SLI_PKT_CTL);
/* 68XX specific */
- max_oqs = CFG_GET_OQ_MAX_Q(CHIP_FIELD(oct, cn6xxx, conf));
+ max_oqs = CFG_GET_OQ_MAX_Q(CHIP_CONF(oct, cn6xxx));
tx_pipe = octeon_read_csr64(oct, CN68XX_SLI_TX_PIPE);
tx_pipe &= 0xffffffffff00ffffULL; /* clear out NUMP field */
tx_pipe |= max_oqs << 16; /* put max_oqs in NUMP field */
diff --git a/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c b/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c
index c7c39a0..e233796 100644
--- a/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c
+++ b/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c
@@ -70,7 +70,6 @@ enum {
INTERFACE_MODE_MIXED,
};
-#define ARRAY_LENGTH(a) (sizeof(a) / sizeof((a)[0]))
#define OCT_ETHTOOL_REGDUMP_LEN 4096
#define OCT_ETHTOOL_REGDUMP_LEN_23XX (4096 * 11)
#define OCT_ETHTOOL_REGSVER 1
@@ -255,14 +254,14 @@ static int lio_get_settings(struct net_device *netdev, struct ethtool_cmd *ecmd)
u32 max_rx = 0, max_tx = 0, tx_count = 0, rx_count = 0;
if (OCTEON_CN6XXX(oct)) {
- struct octeon_config *conf6x = CHIP_FIELD(oct, cn6xxx, conf);
+ struct octeon_config *conf6x = CHIP_CONF(oct, cn6xxx);
max_rx = CFG_GET_OQ_MAX_Q(conf6x);
max_tx = CFG_GET_IQ_MAX_Q(conf6x);
rx_count = CFG_GET_NUM_RXQS_NIC_IF(conf6x, lio->ifidx);
tx_count = CFG_GET_NUM_TXQS_NIC_IF(conf6x, lio->ifidx);
} else if (OCTEON_CN23XX_PF(oct)) {
- struct octeon_config *conf23 = CHIP_FIELD(oct, cn23xx_pf, conf);
+ struct octeon_config *conf23 = CHIP_CONF(oct, cn23xx_pf);
max_rx = CFG_GET_OQ_MAX_Q(conf23);
max_tx = CFG_GET_IQ_MAX_Q(conf23);
@@ -585,14 +584,14 @@ static int lio_set_phys_id(struct net_device *netdev,
rx_pending = 0;
if (OCTEON_CN6XXX(oct)) {
- struct octeon_config *conf6x = CHIP_FIELD(oct, cn6xxx, conf);
+ struct octeon_config *conf6x = CHIP_CONF(oct, cn6xxx);
tx_max_pending = CN6XXX_MAX_IQ_DESCRIPTORS;
rx_max_pending = CN6XXX_MAX_OQ_DESCRIPTORS;
rx_pending = CFG_GET_NUM_RX_DESCS_NIC_IF(conf6x, lio->ifidx);
tx_pending = CFG_GET_NUM_TX_DESCS_NIC_IF(conf6x, lio->ifidx);
} else if (OCTEON_CN23XX_PF(oct)) {
- struct octeon_config *conf23 = CHIP_FIELD(oct, cn23xx_pf, conf);
+ struct octeon_config *conf23 = CHIP_CONF(oct, cn23xx_pf);
tx_max_pending = CN23XX_MAX_IQ_DESCRIPTORS;
rx_max_pending = CN23XX_MAX_OQ_DESCRIPTORS;
diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c
index bf0373a..05612d2 100644
--- a/drivers/net/ethernet/cavium/liquidio/lio_main.c
+++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c
@@ -48,9 +48,6 @@
#define DEFAULT_MSG_ENABLE (NETIF_MSG_DRV | NETIF_MSG_PROBE | NETIF_MSG_LINK)
-#define INCR_INSTRQUEUE_PKT_COUNT(octeon_dev_ptr, iq_no, field, count) \
- (octeon_dev_ptr->instr_queue[iq_no]->stats.field += count)
-
static int debug = -1;
module_param(debug, int, 0644);
MODULE_PARM_DESC(debug, "NETIF_MSG debug bits");
@@ -59,10 +56,6 @@
module_param_string(fw_type, fw_type, sizeof(fw_type), 0000);
MODULE_PARM_DESC(fw_type, "Type of firmware to be loaded. Default \"nic\"");
-static int conf_type;
-module_param(conf_type, int, 0);
-MODULE_PARM_DESC(conf_type, "select octeon configuration 0 default 1 ovs");
-
static int ptp_enable = 1;
/* Bit mask values for lio->ifstate */
@@ -3720,7 +3713,7 @@ static int liquidio_set_vf_link_state(struct net_device *netdev, int vfidx,
return 0;
}
-static struct net_device_ops lionetdevops = {
+static const struct net_device_ops lionetdevops = {
.ndo_open = liquidio_open,
.ndo_stop = liquidio_stop,
.ndo_start_xmit = liquidio_xmit,
@@ -3741,6 +3734,7 @@ static int liquidio_set_vf_link_state(struct net_device *netdev, int vfidx,
.ndo_set_vf_vlan = liquidio_set_vf_vlan,
.ndo_get_vf_config = liquidio_get_vf_config,
.ndo_set_vf_link_state = liquidio_set_vf_link_state,
+ .ndo_select_queue = select_q
};
/** \brief Entry point for the liquidio module
@@ -3752,7 +3746,7 @@ static int __init liquidio_init(void)
init_completion(&first_stage);
- octeon_init_device_list(conf_type);
+ octeon_init_device_list(OCTEON_CONFIG_TYPE_DEFAULT);
if (liquidio_init_pci())
return -EINVAL;
@@ -3973,9 +3967,6 @@ static int setup_nic_devices(struct octeon_device *octeon_dev)
SET_NETDEV_DEV(netdev, &octeon_dev->pci_dev->dev);
- if (num_iqueues > 1)
- lionetdevops.ndo_select_queue = select_q;
-
/* Associate the routines that will handle different
* netdev tasks.
*/
diff --git a/drivers/net/ethernet/cavium/liquidio/liquidio_common.h b/drivers/net/ethernet/cavium/liquidio/liquidio_common.h
index 82a4eed..f308ee4 100644
--- a/drivers/net/ethernet/cavium/liquidio/liquidio_common.h
+++ b/drivers/net/ethernet/cavium/liquidio/liquidio_common.h
@@ -66,7 +66,7 @@ enum octeon_tag_type {
/* Subcodes are used by host driver/apps to identify the sub-operation
* for the core. They only need to by unique for a given subsystem.
*/
-#define OPCODE_SUBCODE(op, sub) (((op & 0x0f) << 8) | ((sub) & 0x7f))
+#define OPCODE_SUBCODE(op, sub) ((((op) & 0x0f) << 8) | ((sub) & 0x7f))
/** OPCODE_CORE subcodes. For future use. */
@@ -89,10 +89,6 @@ enum octeon_tag_type {
#define CORE_DRV_TEST_SCATTER_OP 0xFFF5
-#define OPCODE_SLOW_PATH(rh) \
- (OPCODE_SUBCODE(rh->r.opcode, rh->r.subcode) != \
- OPCODE_SUBCODE(OPCODE_NIC, OPCODE_NIC_NW_DATA))
-
/* Application codes advertised by the core driver initialization packet. */
#define CVM_DRV_APP_START 0x0
#define CVM_DRV_NO_APP 0
@@ -102,31 +98,15 @@ enum octeon_tag_type {
#define CVM_DRV_INVALID_APP (CVM_DRV_APP_START + 0x2)
#define CVM_DRV_APP_END (CVM_DRV_INVALID_APP - 1)
-/* Macro to increment index.
- * Index is incremented by count; if the sum exceeds
- * max, index is wrapped-around to the start.
- */
-#define INCR_INDEX(index, count, max) \
-do { \
- if (((index) + (count)) >= (max)) \
- index = ((index) + (count)) - (max); \
- else \
- index += (count); \
-} while (0)
-
-#define INCR_INDEX_BY1(index, max) \
-do { \
- if ((++(index)) == (max)) \
- index = 0; \
-} while (0)
-
-#define DECR_INDEX(index, count, max) \
-do { \
- if ((count) > (index)) \
- index = ((max) - ((count - index))); \
- else \
- index -= count; \
-} while (0)
+static inline u32 incr_index(u32 index, u32 count, u32 max)
+{
+ if ((index + count) >= max)
+ index = index + count - max;
+ else
+ index += count;
+
+ return index;
+}
#define OCT_BOARD_NAME 32
#define OCT_SERIAL_LEN 64
@@ -827,6 +807,16 @@ struct oct_link_stats {
};
+static inline int opcode_slow_path(union octeon_rh *rh)
+{
+ u16 subcode1, subcode2;
+
+ subcode1 = OPCODE_SUBCODE((rh)->r.opcode, (rh)->r.subcode);
+ subcode2 = OPCODE_SUBCODE(OPCODE_NIC, OPCODE_NIC_NW_DATA);
+
+ return (subcode2 != subcode1);
+}
+
#define LIO68XX_LED_CTRL_ADDR 0x3501
#define LIO68XX_LED_CTRL_CFGON 0x1f
#define LIO68XX_LED_CTRL_CFGOFF 0x100
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_console.c b/drivers/net/ethernet/cavium/liquidio/octeon_console.c
index c0089ed..3265e0b 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_console.c
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_console.c
@@ -138,46 +138,6 @@ struct octeon_pci_console_desc {
};
/**
- * This macro returns the size of a member of a structure.
- * Logically it is the same as "sizeof(s::field)" in C++, but
- * C lacks the "::" operator.
- */
-#define SIZEOF_FIELD(s, field) sizeof(((s *)NULL)->field)
-
-/**
- * This macro returns a member of the cvmx_bootmem_desc
- * structure. These members can't be directly addressed as
- * they might be in memory not directly reachable. In the case
- * where bootmem is compiled with LINUX_HOST, the structure
- * itself might be located on a remote Octeon. The argument
- * "field" is the member name of the cvmx_bootmem_desc to read.
- * Regardless of the type of the field, the return type is always
- * a u64.
- */
-#define CVMX_BOOTMEM_DESC_GET_FIELD(oct, field) \
- __cvmx_bootmem_desc_get(oct, oct->bootmem_desc_addr, \
- offsetof(struct cvmx_bootmem_desc, field), \
- SIZEOF_FIELD(struct cvmx_bootmem_desc, field))
-
-#define __cvmx_bootmem_lock(flags) (flags = flags)
-#define __cvmx_bootmem_unlock(flags) (flags = flags)
-
-/**
- * This macro returns a member of the
- * cvmx_bootmem_named_block_desc structure. These members can't
- * be directly addressed as they might be in memory not directly
- * reachable. In the case where bootmem is compiled with
- * LINUX_HOST, the structure itself might be located on a remote
- * Octeon. The argument "field" is the member name of the
- * cvmx_bootmem_named_block_desc to read. Regardless of the type
- * of the field, the return type is always a u64. The "addr"
- * parameter is the physical address of the structure.
- */
-#define CVMX_BOOTMEM_NAMED_GET_FIELD(oct, addr, field) \
- __cvmx_bootmem_desc_get(oct, addr, \
- offsetof(struct cvmx_bootmem_named_block_desc, field), \
- SIZEOF_FIELD(struct cvmx_bootmem_named_block_desc, field))
-/**
* \brief determines if a given console has debug enabled.
* @param console console to check
* @returns 1 = enabled. 0 otherwise
@@ -258,10 +218,15 @@ static int __cvmx_bootmem_check_version(struct octeon_device *oct,
oct->bootmem_desc_addr =
octeon_read_device_mem64(oct,
BOOTLOADER_PCI_READ_DESC_ADDR);
- major_version =
- (u32)CVMX_BOOTMEM_DESC_GET_FIELD(oct, major_version);
- minor_version =
- (u32)CVMX_BOOTMEM_DESC_GET_FIELD(oct, minor_version);
+ major_version = (u32)__cvmx_bootmem_desc_get(
+ oct, oct->bootmem_desc_addr,
+ offsetof(struct cvmx_bootmem_desc, major_version),
+ FIELD_SIZEOF(struct cvmx_bootmem_desc, major_version));
+ minor_version = (u32)__cvmx_bootmem_desc_get(
+ oct, oct->bootmem_desc_addr,
+ offsetof(struct cvmx_bootmem_desc, minor_version),
+ FIELD_SIZEOF(struct cvmx_bootmem_desc, minor_version));
+
dev_dbg(&oct->pci_dev->dev, "%s: major_version=%d\n", __func__,
major_version);
if ((major_version > 3) ||
@@ -284,10 +249,20 @@ static int __cvmx_bootmem_check_version(struct octeon_device *oct,
u64 named_addr = cvmx_bootmem_phy_named_block_find(oct, name, flags);
if (named_addr) {
- desc->base_addr = CVMX_BOOTMEM_NAMED_GET_FIELD(oct, named_addr,
- base_addr);
- desc->size =
- CVMX_BOOTMEM_NAMED_GET_FIELD(oct, named_addr, size);
+ desc->base_addr = __cvmx_bootmem_desc_get(
+ oct, named_addr,
+ offsetof(struct cvmx_bootmem_named_block_desc,
+ base_addr),
+ FIELD_SIZEOF(
+ struct cvmx_bootmem_named_block_desc,
+ base_addr));
+ desc->size = __cvmx_bootmem_desc_get(oct, named_addr,
+ offsetof(struct cvmx_bootmem_named_block_desc,
+ size),
+ FIELD_SIZEOF(
+ struct cvmx_bootmem_named_block_desc,
+ size));
+
strncpy(desc->name, name, sizeof(desc->name));
desc->name[sizeof(desc->name) - 1] = 0;
return &oct->bootmem_named_block_desc;
@@ -302,22 +277,41 @@ static u64 cvmx_bootmem_phy_named_block_find(struct octeon_device *oct,
{
u64 result = 0;
- __cvmx_bootmem_lock(flags);
if (!__cvmx_bootmem_check_version(oct, 3)) {
u32 i;
- u64 named_block_array_addr =
- CVMX_BOOTMEM_DESC_GET_FIELD(oct,
- named_block_array_addr);
- u32 num_blocks = (u32)
- CVMX_BOOTMEM_DESC_GET_FIELD(oct, nb_num_blocks);
- u32 name_length = (u32)
- CVMX_BOOTMEM_DESC_GET_FIELD(oct, named_block_name_len);
+
+ u64 named_block_array_addr = __cvmx_bootmem_desc_get(
+ oct, oct->bootmem_desc_addr,
+ offsetof(struct cvmx_bootmem_desc,
+ named_block_array_addr),
+ FIELD_SIZEOF(struct cvmx_bootmem_desc,
+ named_block_array_addr));
+ u32 num_blocks = (u32)__cvmx_bootmem_desc_get(
+ oct, oct->bootmem_desc_addr,
+ offsetof(struct cvmx_bootmem_desc,
+ nb_num_blocks),
+ FIELD_SIZEOF(struct cvmx_bootmem_desc,
+ nb_num_blocks));
+
+ u32 name_length = (u32)__cvmx_bootmem_desc_get(
+ oct, oct->bootmem_desc_addr,
+ offsetof(struct cvmx_bootmem_desc,
+ named_block_name_len),
+ FIELD_SIZEOF(struct cvmx_bootmem_desc,
+ named_block_name_len));
+
u64 named_addr = named_block_array_addr;
for (i = 0; i < num_blocks; i++) {
- u64 named_size =
- CVMX_BOOTMEM_NAMED_GET_FIELD(oct, named_addr,
- size);
+ u64 named_size = __cvmx_bootmem_desc_get(
+ oct, named_addr,
+ offsetof(
+ struct cvmx_bootmem_named_block_desc,
+ size),
+ FIELD_SIZEOF(
+ struct cvmx_bootmem_named_block_desc,
+ size));
+
if (name && named_size) {
char *name_tmp =
kmalloc(name_length + 1, GFP_KERNEL);
@@ -342,7 +336,6 @@ static u64 cvmx_bootmem_phy_named_block_find(struct octeon_device *oct,
sizeof(struct cvmx_bootmem_named_block_desc);
}
}
- __cvmx_bootmem_unlock(flags);
return result;
}
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_device.c b/drivers/net/ethernet/cavium/liquidio/octeon_device.c
index 4a30d7c..79c8875 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_device.c
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_device.c
@@ -516,11 +516,6 @@
}
};
-enum {
- OCTEON_CONFIG_TYPE_DEFAULT = 0,
- NUM_OCTEON_CONFS,
-};
-
static struct octeon_config_ptr {
u32 conf_type;
} oct_conf_info[MAX_OCTEON_DEVICES] = {
@@ -792,10 +787,9 @@ int octeon_setup_instr_queues(struct octeon_device *oct)
if (OCTEON_CN6XXX(oct))
num_descs =
- CFG_GET_NUM_DEF_TX_DESCS(CHIP_FIELD(oct, cn6xxx, conf));
+ CFG_GET_NUM_DEF_TX_DESCS(CHIP_CONF(oct, cn6xxx));
else if (OCTEON_CN23XX_PF(oct))
- num_descs = CFG_GET_NUM_DEF_TX_DESCS(CHIP_FIELD(oct, cn23xx_pf,
- conf));
+ num_descs = CFG_GET_NUM_DEF_TX_DESCS(CHIP_CONF(oct, cn23xx_pf));
oct->num_iqs = 0;
@@ -835,14 +829,12 @@ int octeon_setup_output_queues(struct octeon_device *oct)
if (OCTEON_CN6XXX(oct)) {
num_descs =
- CFG_GET_NUM_DEF_RX_DESCS(CHIP_FIELD(oct, cn6xxx, conf));
+ CFG_GET_NUM_DEF_RX_DESCS(CHIP_CONF(oct, cn6xxx));
desc_size =
- CFG_GET_DEF_RX_BUF_SIZE(CHIP_FIELD(oct, cn6xxx, conf));
+ CFG_GET_DEF_RX_BUF_SIZE(CHIP_CONF(oct, cn6xxx));
} else if (OCTEON_CN23XX_PF(oct)) {
- num_descs = CFG_GET_NUM_DEF_RX_DESCS(CHIP_FIELD(oct, cn23xx_pf,
- conf));
- desc_size = CFG_GET_DEF_RX_BUF_SIZE(CHIP_FIELD(oct, cn23xx_pf,
- conf));
+ num_descs = CFG_GET_NUM_DEF_RX_DESCS(CHIP_CONF(oct, cn23xx_pf));
+ desc_size = CFG_GET_DEF_RX_BUF_SIZE(CHIP_CONF(oct, cn23xx_pf));
}
oct->num_oqs = 0;
oct->droq[0] = vmalloc_node(sizeof(*oct->droq[0]), numa_node);
@@ -1071,10 +1063,10 @@ int octeon_core_drv_init(struct octeon_recv_info *recv_info, void *buf)
if (OCTEON_CN6XXX(oct))
num_nic_ports =
- CFG_GET_NUM_NIC_PORTS(CHIP_FIELD(oct, cn6xxx, conf));
+ CFG_GET_NUM_NIC_PORTS(CHIP_CONF(oct, cn6xxx));
else if (OCTEON_CN23XX_PF(oct))
num_nic_ports =
- CFG_GET_NUM_NIC_PORTS(CHIP_FIELD(oct, cn23xx_pf, conf));
+ CFG_GET_NUM_NIC_PORTS(CHIP_CONF(oct, cn23xx_pf));
if (atomic_read(&oct->status) >= OCT_DEV_RUNNING) {
dev_err(&oct->pci_dev->dev, "Received CORE OK when device state is 0x%x\n",
@@ -1169,10 +1161,10 @@ struct octeon_config *octeon_get_conf(struct octeon_device *oct)
if (OCTEON_CN6XXX(oct)) {
default_oct_conf =
- (struct octeon_config *)(CHIP_FIELD(oct, cn6xxx, conf));
+ (struct octeon_config *)(CHIP_CONF(oct, cn6xxx));
} else if (OCTEON_CN23XX_PF(oct)) {
default_oct_conf = (struct octeon_config *)
- (CHIP_FIELD(oct, cn23xx_pf, conf));
+ (CHIP_CONF(oct, cn23xx_pf));
}
return default_oct_conf;
}
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_device.h b/drivers/net/ethernet/cavium/liquidio/octeon_device.h
index c59b77e..5ce2048 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_device.h
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_device.h
@@ -48,6 +48,11 @@ enum octeon_pci_swap_mode {
OCTEON_PCI_32BIT_LW_SWAP = 3
};
+enum {
+ OCTEON_CONFIG_TYPE_DEFAULT = 0,
+ NUM_OCTEON_CONFS,
+};
+
#define OCTEON_OUTPUT_INTR (2)
#define OCTEON_MBOX_INTR (4)
#define OCTEON_ALL_INTR 0xff
@@ -524,12 +529,14 @@ struct octeon_device {
#define OCT_DRV_ONLINE 1
#define OCT_DRV_OFFLINE 2
-#define OCTEON_CN6XXX(oct) ((oct->chip_id == OCTEON_CN66XX) || \
- (oct->chip_id == OCTEON_CN68XX))
-#define OCTEON_CN23XX_PF(oct) (oct->chip_id == OCTEON_CN23XX_PF_VID)
+#define OCTEON_CN6XXX(oct) ({ \
+ typeof(oct) _oct = (oct); \
+ ((_oct->chip_id == OCTEON_CN66XX) || \
+ (_oct->chip_id == OCTEON_CN68XX)); })
+#define OCTEON_CN23XX_PF(oct) ((oct)->chip_id == OCTEON_CN23XX_PF_VID)
#define OCTEON_CN23XX_VF(oct) ((oct)->chip_id == OCTEON_CN23XX_VF_VID)
-#define CHIP_FIELD(oct, TYPE, field) \
- (((struct octeon_ ## TYPE *)(oct->chip))->field)
+#define CHIP_CONF(oct, TYPE) \
+ (((struct octeon_ ## TYPE *)((oct)->chip))->conf)
struct oct_intrmod_cmd {
struct octeon_device *oct_dev;
@@ -641,16 +648,16 @@ static inline u16 OCTEON_MINOR_REV(struct octeon_device *oct)
/* Routines for reading and writing CSRs */
#define octeon_write_csr(oct_dev, reg_off, value) \
- writel(value, oct_dev->mmio[0].hw_addr + reg_off)
+ writel(value, (oct_dev)->mmio[0].hw_addr + (reg_off))
#define octeon_write_csr64(oct_dev, reg_off, val64) \
- writeq(val64, oct_dev->mmio[0].hw_addr + reg_off)
+ writeq(val64, (oct_dev)->mmio[0].hw_addr + (reg_off))
#define octeon_read_csr(oct_dev, reg_off) \
- readl(oct_dev->mmio[0].hw_addr + reg_off)
+ readl((oct_dev)->mmio[0].hw_addr + (reg_off))
#define octeon_read_csr64(oct_dev, reg_off) \
- readq(oct_dev->mmio[0].hw_addr + reg_off)
+ readq((oct_dev)->mmio[0].hw_addr + (reg_off))
/**
* Checks if memory access is okay
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_droq.c b/drivers/net/ethernet/cavium/liquidio/octeon_droq.c
index d7ba000..8bf1ac76 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_droq.c
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_droq.c
@@ -29,9 +29,6 @@
#include "cn66xx_device.h"
#include "cn23xx_pf_device.h"
-#define CVM_MIN(d1, d2) (((d1) < (d2)) ? (d1) : (d2))
-#define CVM_MAX(d1, d2) (((d1) > (d2)) ? (d1) : (d2))
-
struct niclist {
struct list_head list;
void *ptr;
@@ -254,13 +251,13 @@ int octeon_init_droq(struct octeon_device *oct,
c_num_descs = num_descs;
c_buf_size = desc_size;
if (OCTEON_CN6XXX(oct)) {
- struct octeon_config *conf6x = CHIP_FIELD(oct, cn6xxx, conf);
+ struct octeon_config *conf6x = CHIP_CONF(oct, cn6xxx);
c_pkts_per_intr = (u32)CFG_GET_OQ_PKTS_PER_INTR(conf6x);
c_refill_threshold =
(u32)CFG_GET_OQ_REFILL_THRESHOLD(conf6x);
} else if (OCTEON_CN23XX_PF(oct)) {
- struct octeon_config *conf23 = CHIP_FIELD(oct, cn23xx_pf, conf);
+ struct octeon_config *conf23 = CHIP_CONF(oct, cn23xx_pf);
c_pkts_per_intr = (u32)CFG_GET_OQ_PKTS_PER_INTR(conf23);
c_refill_threshold = (u32)CFG_GET_OQ_REFILL_THRESHOLD(conf23);
@@ -405,7 +402,7 @@ static inline struct octeon_recv_info *octeon_create_recv_info(
recv_pkt->buffer_ptr[i] = droq->recv_buf_list[idx].buffer;
droq->recv_buf_list[idx].buffer = NULL;
- INCR_INDEX_BY1(idx, droq->max_count);
+ idx = incr_index(idx, 1, droq->max_count);
bytes_left -= droq->buffer_size;
i++;
buf_cnt--;
@@ -436,14 +433,15 @@ static inline struct octeon_recv_info *octeon_create_recv_info(
droq->recv_buf_list[refill_index].buffer = NULL;
desc_ring[refill_index].buffer_ptr = 0;
do {
- INCR_INDEX_BY1(droq->refill_idx,
- droq->max_count);
+ droq->refill_idx = incr_index(droq->refill_idx,
+ 1,
+ droq->max_count);
desc_refilled++;
droq->refill_count--;
} while (droq->recv_buf_list[droq->refill_idx].
buffer);
}
- INCR_INDEX_BY1(refill_index, droq->max_count);
+ refill_index = incr_index(refill_index, 1, droq->max_count);
} /* while */
return desc_refilled;
}
@@ -510,7 +508,8 @@ static inline struct octeon_recv_info *octeon_create_recv_info(
/* Reset any previous values in the length field. */
droq->info_list[droq->refill_idx].length = 0;
- INCR_INDEX_BY1(droq->refill_idx, droq->max_count);
+ droq->refill_idx = incr_index(droq->refill_idx, 1,
+ droq->max_count);
desc_refilled++;
droq->refill_count--;
}
@@ -595,7 +594,8 @@ static inline void octeon_droq_drop_packets(struct octeon_device *oct,
buf_cnt = 1;
}
- INCR_INDEX(droq->read_idx, buf_cnt, droq->max_count);
+ droq->read_idx = incr_index(droq->read_idx, buf_cnt,
+ droq->max_count);
droq->refill_count += buf_cnt;
}
}
@@ -635,11 +635,12 @@ static inline void octeon_droq_drop_packets(struct octeon_device *oct,
rh = &info->rh;
total_len += (u32)info->length;
- if (OPCODE_SLOW_PATH(rh)) {
+ if (opcode_slow_path(rh)) {
u32 buf_cnt;
buf_cnt = octeon_droq_dispatch_pkt(oct, droq, rh, info);
- INCR_INDEX(droq->read_idx, buf_cnt, droq->max_count);
+ droq->read_idx = incr_index(droq->read_idx,
+ buf_cnt, droq->max_count);
droq->refill_count += buf_cnt;
} else {
if (info->length <= droq->buffer_size) {
@@ -653,7 +654,8 @@ static inline void octeon_droq_drop_packets(struct octeon_device *oct,
droq->recv_buf_list[droq->read_idx].buffer =
NULL;
- INCR_INDEX_BY1(droq->read_idx, droq->max_count);
+ droq->read_idx = incr_index(droq->read_idx, 1,
+ droq->max_count);
droq->refill_count++;
} else {
nicbuf = octeon_fast_packet_alloc((u32)
@@ -685,8 +687,9 @@ static inline void octeon_droq_drop_packets(struct octeon_device *oct,
}
pkt_len += cpy_len;
- INCR_INDEX_BY1(droq->read_idx,
- droq->max_count);
+ droq->read_idx =
+ incr_index(droq->read_idx, 1,
+ droq->max_count);
droq->refill_count++;
}
}
@@ -800,9 +803,8 @@ static inline void octeon_droq_drop_packets(struct octeon_device *oct,
while (total_pkts_processed < budget) {
octeon_droq_check_hw_for_pkts(droq);
- pkts_available =
- CVM_MIN((budget - total_pkts_processed),
- (u32)(atomic_read(&droq->pkts_pending)));
+ pkts_available = min((budget - total_pkts_processed),
+ (u32)(atomic_read(&droq->pkts_pending)));
if (pkts_available == 0)
break;
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_iq.h b/drivers/net/ethernet/cavium/liquidio/octeon_iq.h
index c67152a..e04ca8f 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_iq.h
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_iq.h
@@ -303,6 +303,9 @@ struct octeon_sc_buffer_pool {
atomic_t alloc_buf_count;
};
+#define INCR_INSTRQUEUE_PKT_COUNT(octeon_dev_ptr, iq_no, field, count) \
+ (((octeon_dev_ptr)->instr_queue[iq_no]->stats.field) += count)
+
int octeon_setup_sc_buffer_pool(struct octeon_device *oct);
int octeon_free_sc_buffer_pool(struct octeon_device *oct);
struct octeon_soft_command *
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_mem_ops.c b/drivers/net/ethernet/cavium/liquidio/octeon_mem_ops.c
index 98d8066d..13a18c9 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_mem_ops.c
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_mem_ops.c
@@ -36,7 +36,7 @@
oct->fn_list.bar1_idx_write(oct, idx, mask);
}
#else
-#define octeon_toggle_bar1_swapmode(oct, idx) (oct = oct)
+#define octeon_toggle_bar1_swapmode(oct, idx)
#endif
static void
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_network.h b/drivers/net/ethernet/cavium/liquidio/octeon_network.h
index 172b70e..e94edc8 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_network.h
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_network.h
@@ -128,7 +128,7 @@ struct lio {
#define LIO_SIZE (sizeof(struct lio))
#define GET_LIO(netdev) ((struct lio *)netdev_priv(netdev))
-#define CIU3_WDOG(c) (0x1010000020000ULL + (c << 3))
+#define CIU3_WDOG(c) (0x1010000020000ULL + ((c) << 3))
#define CIU3_WDOG_MASK 12ULL
#define LIO_MONITOR_WDOG_EXPIRE 1
#define LIO_MONITOR_CORE_STUCK_MSGD 2
@@ -339,9 +339,9 @@ static inline void tx_buffer_free(void *buffer)
}
#define lio_dma_alloc(oct, size, dma_addr) \
- dma_alloc_coherent(&oct->pci_dev->dev, size, dma_addr, GFP_KERNEL)
+ dma_alloc_coherent(&(oct)->pci_dev->dev, size, dma_addr, GFP_KERNEL)
#define lio_dma_free(oct, size, virt_addr, dma_addr) \
- dma_free_coherent(&oct->pci_dev->dev, size, virt_addr, dma_addr)
+ dma_free_coherent(&(oct)->pci_dev->dev, size, virt_addr, dma_addr)
static inline
void *get_rbd(struct sk_buff *skb)
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_nic.h b/drivers/net/ethernet/cavium/liquidio/octeon_nic.h
index b059979..0c7a5c9 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_nic.h
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_nic.h
@@ -64,7 +64,7 @@ struct octnic_ctrl_pkt {
octnic_ctrl_pkt_cb_fn_t cb_fn;
};
-#define MAX_UDD_SIZE(nctrl) (sizeof(nctrl->udd))
+#define MAX_UDD_SIZE(nctrl) (sizeof((nctrl)->udd))
/** Structure of data information passed by the NIC module to the OSI
* layer when forwarding data to Octeon device software.
diff --git a/drivers/net/ethernet/cavium/liquidio/request_manager.c b/drivers/net/ethernet/cavium/liquidio/request_manager.c
index 0cc1583..8531a00 100644
--- a/drivers/net/ethernet/cavium/liquidio/request_manager.c
+++ b/drivers/net/ethernet/cavium/liquidio/request_manager.c
@@ -29,9 +29,6 @@
#include "cn66xx_device.h"
#include "cn23xx_pf_device.h"
-#define INCR_INSTRQUEUE_PKT_COUNT(octeon_dev_ptr, iq_no, field, count) \
- (octeon_dev_ptr->instr_queue[iq_no]->stats.field += count)
-
struct iq_post_status {
int status;
int index;
@@ -68,9 +65,9 @@ int octeon_init_instr_queue(struct octeon_device *oct,
int numa_node = cpu_to_node(iq_no % num_online_cpus());
if (OCTEON_CN6XXX(oct))
- conf = &(CFG_GET_IQ_CFG(CHIP_FIELD(oct, cn6xxx, conf)));
+ conf = &(CFG_GET_IQ_CFG(CHIP_CONF(oct, cn6xxx)));
else if (OCTEON_CN23XX_PF(oct))
- conf = &(CFG_GET_IQ_CFG(CHIP_FIELD(oct, cn23xx_pf, conf)));
+ conf = &(CFG_GET_IQ_CFG(CHIP_CONF(oct, cn23xx_pf)));
if (!conf) {
dev_err(&oct->pci_dev->dev, "Unsupported Chip %x\n",
oct->chip_id);
@@ -182,10 +179,10 @@ int octeon_delete_instr_queue(struct octeon_device *oct, u32 iq_no)
if (OCTEON_CN6XXX(oct))
desc_size =
- CFG_GET_IQ_INSTR_TYPE(CHIP_FIELD(oct, cn6xxx, conf));
+ CFG_GET_IQ_INSTR_TYPE(CHIP_CONF(oct, cn6xxx));
else if (OCTEON_CN23XX_PF(oct))
desc_size =
- CFG_GET_IQ_INSTR_TYPE(CHIP_FIELD(oct, cn23xx_pf, conf));
+ CFG_GET_IQ_INSTR_TYPE(CHIP_CONF(oct, cn23xx_pf));
vfree(iq->request_list);
@@ -317,7 +314,8 @@ static inline void __copy_cmd_into_iq(struct octeon_instr_queue *iq,
/* "index" is returned, host_write_index is modified. */
st.index = iq->host_write_index;
- INCR_INDEX_BY1(iq->host_write_index, iq->max_count);
+ iq->host_write_index = incr_index(iq->host_write_index, 1,
+ iq->max_count);
iq->fill_cnt++;
/* Flush the command into memory. We need to be sure the data is in
@@ -432,7 +430,7 @@ static inline void __copy_cmd_into_iq(struct octeon_instr_queue *iq,
skip_this:
inst_count++;
- INCR_INDEX_BY1(old, iq->max_count);
+ old = incr_index(old, 1, iq->max_count);
if ((napi_budget) && (inst_count >= napi_budget))
break;
--
1.8.3.1
^ permalink raw reply related
* [PATCH net-next 0/2] sfc: enable 4-tuple UDP RSS hashing
From: Edward Cree @ 2016-11-03 22:10 UTC (permalink / raw)
To: netdev, davem, linux-net-drivers
EF10 based NICs have configurable RSS hash fields, and can be made to take the
ports into the hash on UDP (they already do so for TCP). This patch series
enables this, in order to improve spreading of UDP traffic.
Edward Cree (2):
sfc: enable 4-tuple RSS hashing for UDP
sfc: report 4-tuple UDP hashing to ethtool, if it's enabled
drivers/net/ethernet/sfc/ef10.c | 84 +++++++++++++++++++++++++++++++++++
drivers/net/ethernet/sfc/ethtool.c | 12 +++--
drivers/net/ethernet/sfc/net_driver.h | 2 +
3 files changed, 94 insertions(+), 4 deletions(-)
^ permalink raw reply
* [PATCH net-next 1/2] sfc: enable 4-tuple RSS hashing for UDP
From: Edward Cree @ 2016-11-03 22:12 UTC (permalink / raw)
To: linux-net-drivers, netdev, davem
In-Reply-To: <ba5aa0bc-277b-5b47-89f8-fad1010b77f9@solarflare.com>
This improves UDP spreading, and also slightly improves GRO performance
of encapsulated TCP on 7000 series NICs.
Signed-off-by: Edward Cree <ecree@solarflare.com>
---
drivers/net/ethernet/sfc/ef10.c | 82 +++++++++++++++++++++++++++++++++++++++++
1 file changed, 82 insertions(+)
diff --git a/drivers/net/ethernet/sfc/ef10.c b/drivers/net/ethernet/sfc/ef10.c
index 00279da..9f6d769 100644
--- a/drivers/net/ethernet/sfc/ef10.c
+++ b/drivers/net/ethernet/sfc/ef10.c
@@ -2245,6 +2245,84 @@ static void efx_ef10_tx_write(struct efx_tx_queue *tx_queue)
}
}
+#define RSS_MODE_HASH_ADDRS (1 << RSS_MODE_HASH_SRC_ADDR_LBN |\
+ 1 << RSS_MODE_HASH_DST_ADDR_LBN)
+#define RSS_MODE_HASH_PORTS (1 << RSS_MODE_HASH_SRC_PORT_LBN |\
+ 1 << RSS_MODE_HASH_DST_PORT_LBN)
+#define RSS_CONTEXT_FLAGS_DEFAULT (1 << MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_TOEPLITZ_IPV4_EN_LBN |\
+ 1 << MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_TOEPLITZ_TCPV4_EN_LBN |\
+ 1 << MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_TOEPLITZ_IPV6_EN_LBN |\
+ 1 << MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_TOEPLITZ_TCPV6_EN_LBN |\
+ (RSS_MODE_HASH_ADDRS | RSS_MODE_HASH_PORTS) << MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_TCP_IPV4_RSS_MODE_LBN |\
+ RSS_MODE_HASH_ADDRS << MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_UDP_IPV4_RSS_MODE_LBN |\
+ RSS_MODE_HASH_ADDRS << MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_OTHER_IPV4_RSS_MODE_LBN |\
+ (RSS_MODE_HASH_ADDRS | RSS_MODE_HASH_PORTS) << MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_TCP_IPV6_RSS_MODE_LBN |\
+ RSS_MODE_HASH_ADDRS << MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_UDP_IPV6_RSS_MODE_LBN |\
+ RSS_MODE_HASH_ADDRS << MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_OTHER_IPV6_RSS_MODE_LBN)
+
+static int efx_ef10_get_rss_flags(struct efx_nic *efx, u32 context, u32 *flags)
+{
+ /* Firmware had a bug (sfc bug 61952) where it would not actually
+ * fill in the flags field in the response to MC_CMD_RSS_CONTEXT_GET_FLAGS.
+ * This meant that it would always contain whatever was previously
+ * in the MCDI buffer. Fortunately, all firmware versions with
+ * this bug have the same default flags value for a newly-allocated
+ * RSS context, and the only time we want to get the flags is just
+ * after allocating. Moreover, the response has a 32-bit hole
+ * where the context ID would be in the request, so we can use an
+ * overlength buffer in the request and pre-fill the flags field
+ * with what we believe the default to be. Thus if the firmware
+ * has the bug, it will leave our pre-filled value in the flags
+ * field of the response, and we will get the right answer.
+ *
+ * However, this does mean that this function should NOT be used if
+ * the RSS context flags might not be their defaults - it is ONLY
+ * reliably correct for a newly-allocated RSS context.
+ */
+ MCDI_DECLARE_BUF(inbuf, MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_LEN);
+ MCDI_DECLARE_BUF(outbuf, MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_LEN);
+ size_t outlen;
+ int rc;
+
+ /* Check we have a hole for the context ID */
+ BUILD_BUG_ON(MC_CMD_RSS_CONTEXT_GET_FLAGS_IN_LEN != MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_FLAGS_OFST);
+ MCDI_SET_DWORD(inbuf, RSS_CONTEXT_GET_FLAGS_IN_RSS_CONTEXT_ID, context);
+ MCDI_SET_DWORD(inbuf, RSS_CONTEXT_GET_FLAGS_OUT_FLAGS,
+ RSS_CONTEXT_FLAGS_DEFAULT);
+ rc = efx_mcdi_rpc(efx, MC_CMD_RSS_CONTEXT_GET_FLAGS, inbuf,
+ sizeof(inbuf), outbuf, sizeof(outbuf), &outlen);
+ if (rc == 0) {
+ if (outlen < MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_LEN)
+ rc = -EIO;
+ else
+ *flags = MCDI_DWORD(outbuf, RSS_CONTEXT_GET_FLAGS_OUT_FLAGS);
+ }
+ return rc;
+}
+
+/* Attempt to enable 4-tuple UDP hashing on the specified RSS context.
+ * If we fail, we just leave the RSS context at its default hash settings,
+ * which is safe but may slightly reduce performance.
+ * Defaults are 4-tuple for TCP and 2-tuple for UDP and other-IP, so we
+ * just need to set the UDP ports flags (for both IP versions).
+ */
+static void efx_ef10_set_rss_flags(struct efx_nic *efx, u32 context)
+{
+ MCDI_DECLARE_BUF(inbuf, MC_CMD_RSS_CONTEXT_SET_FLAGS_IN_LEN);
+ u32 flags;
+
+ BUILD_BUG_ON(MC_CMD_RSS_CONTEXT_SET_FLAGS_OUT_LEN != 0);
+
+ if (efx_ef10_get_rss_flags(efx, context, &flags) != 0)
+ return;
+ MCDI_SET_DWORD(inbuf, RSS_CONTEXT_SET_FLAGS_IN_RSS_CONTEXT_ID, context);
+ flags |= RSS_MODE_HASH_PORTS << MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_UDP_IPV4_RSS_MODE_LBN;
+ flags |= RSS_MODE_HASH_PORTS << MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_UDP_IPV6_RSS_MODE_LBN;
+ MCDI_SET_DWORD(inbuf, RSS_CONTEXT_SET_FLAGS_IN_FLAGS, flags);
+ efx_mcdi_rpc(efx, MC_CMD_RSS_CONTEXT_SET_FLAGS, inbuf, sizeof(inbuf),
+ NULL, 0, NULL);
+}
+
static int efx_ef10_alloc_rss_context(struct efx_nic *efx, u32 *context,
bool exclusive, unsigned *context_size)
{
@@ -2290,6 +2368,10 @@ static int efx_ef10_alloc_rss_context(struct efx_nic *efx, u32 *context,
if (context_size)
*context_size = rss_spread;
+ if (nic_data->datapath_caps &
+ 1 << MC_CMD_GET_CAPABILITIES_OUT_ADDITIONAL_RSS_MODES_LBN)
+ efx_ef10_set_rss_flags(efx, *context);
+
return 0;
}
^ permalink raw reply related
* [PATCH net-next 2/2] sfc: report 4-tuple UDP hashing to ethtool, if it's enabled
From: Edward Cree @ 2016-11-03 22:12 UTC (permalink / raw)
To: linux-net-drivers, netdev, davem
In-Reply-To: <ba5aa0bc-277b-5b47-89f8-fad1010b77f9@solarflare.com>
Signed-off-by: Edward Cree <ecree@solarflare.com>
---
drivers/net/ethernet/sfc/ef10.c | 6 ++++--
drivers/net/ethernet/sfc/ethtool.c | 12 ++++++++----
drivers/net/ethernet/sfc/net_driver.h | 2 ++
3 files changed, 14 insertions(+), 6 deletions(-)
diff --git a/drivers/net/ethernet/sfc/ef10.c b/drivers/net/ethernet/sfc/ef10.c
index 9f6d769..e61807e 100644
--- a/drivers/net/ethernet/sfc/ef10.c
+++ b/drivers/net/ethernet/sfc/ef10.c
@@ -2319,8 +2319,10 @@ static void efx_ef10_set_rss_flags(struct efx_nic *efx, u32 context)
flags |= RSS_MODE_HASH_PORTS << MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_UDP_IPV4_RSS_MODE_LBN;
flags |= RSS_MODE_HASH_PORTS << MC_CMD_RSS_CONTEXT_GET_FLAGS_OUT_UDP_IPV6_RSS_MODE_LBN;
MCDI_SET_DWORD(inbuf, RSS_CONTEXT_SET_FLAGS_IN_FLAGS, flags);
- efx_mcdi_rpc(efx, MC_CMD_RSS_CONTEXT_SET_FLAGS, inbuf, sizeof(inbuf),
- NULL, 0, NULL);
+ if (!efx_mcdi_rpc(efx, MC_CMD_RSS_CONTEXT_SET_FLAGS, inbuf, sizeof(inbuf),
+ NULL, 0, NULL))
+ /* Succeeded, so UDP 4-tuple is now enabled */
+ efx->rx_hash_udp_4tuple = true;
}
static int efx_ef10_alloc_rss_context(struct efx_nic *efx, u32 *context,
diff --git a/drivers/net/ethernet/sfc/ethtool.c b/drivers/net/ethernet/sfc/ethtool.c
index 445ccdb..bf126f9 100644
--- a/drivers/net/ethernet/sfc/ethtool.c
+++ b/drivers/net/ethernet/sfc/ethtool.c
@@ -968,20 +968,24 @@ efx_ethtool_get_rxnfc(struct net_device *net_dev,
info->data = 0;
switch (info->flow_type) {
+ case UDP_V4_FLOW:
+ if (efx->rx_hash_udp_4tuple)
+ /* fall through */
case TCP_V4_FLOW:
- info->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
+ info->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
/* fall through */
- case UDP_V4_FLOW:
case SCTP_V4_FLOW:
case AH_ESP_V4_FLOW:
case IPV4_FLOW:
info->data |= RXH_IP_SRC | RXH_IP_DST;
min_revision = EFX_REV_FALCON_B0;
break;
+ case UDP_V6_FLOW:
+ if (efx->rx_hash_udp_4tuple)
+ /* fall through */
case TCP_V6_FLOW:
- info->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
+ info->data |= RXH_L4_B_0_1 | RXH_L4_B_2_3;
/* fall through */
- case UDP_V6_FLOW:
case SCTP_V6_FLOW:
case AH_ESP_V6_FLOW:
case IPV6_FLOW:
diff --git a/drivers/net/ethernet/sfc/net_driver.h b/drivers/net/ethernet/sfc/net_driver.h
index 99d8c82..fec51c4 100644
--- a/drivers/net/ethernet/sfc/net_driver.h
+++ b/drivers/net/ethernet/sfc/net_driver.h
@@ -853,6 +853,7 @@ struct vfdi_status;
* @rx_hash_key: Toeplitz hash key for RSS
* @rx_indir_table: Indirection table for RSS
* @rx_scatter: Scatter mode enabled for receives
+ * @rx_hash_udp_4tuple: UDP 4-tuple hashing enabled
* @int_error_count: Number of internal errors seen recently
* @int_error_expire: Time at which error count will be expired
* @irq_soft_enabled: Are IRQs soft-enabled? If not, IRQ handler will
@@ -990,6 +991,7 @@ struct efx_nic {
u8 rx_hash_key[40];
u32 rx_indir_table[128];
bool rx_scatter;
+ bool rx_hash_udp_4tuple;
unsigned int_error_count;
unsigned long int_error_expire;
^ permalink raw reply related
* Re: [PATCH net-next RFC WIP] Patch for XDP support for virtio_net
From: John Fastabend @ 2016-11-03 22:20 UTC (permalink / raw)
To: Michael S. Tsirkin, Shrijeet Mukherjee
Cc: Jesper Dangaard Brouer, Thomas Graf, Alexei Starovoitov,
Jakub Kicinski, David Miller, alexander.duyck, shrijeet, tom,
netdev, Roopa Prabhu, Nikolay Aleksandrov
In-Reply-To: <581ADCD1.3040404@gmail.com>
On 16-11-02 11:44 PM, John Fastabend wrote:
> On 16-11-02 09:11 PM, Michael S. Tsirkin wrote:
>> On Wed, Nov 02, 2016 at 06:28:34PM -0700, Shrijeet Mukherjee wrote:
>>>> -----Original Message-----
>>>> From: Jesper Dangaard Brouer [mailto:brouer@redhat.com]
>>>> Sent: Wednesday, November 2, 2016 7:27 AM
>>>> To: Thomas Graf <tgraf@suug.ch>
>>>> Cc: Shrijeet Mukherjee <shm@cumulusnetworks.com>; Alexei Starovoitov
>>>> <alexei.starovoitov@gmail.com>; Jakub Kicinski <kubakici@wp.pl>; John
>>>> Fastabend <john.fastabend@gmail.com>; David Miller
>>>> <davem@davemloft.net>; alexander.duyck@gmail.com; mst@redhat.com;
>>>> shrijeet@gmail.com; tom@herbertland.com; netdev@vger.kernel.org;
>>>> Roopa Prabhu <roopa@cumulusnetworks.com>; Nikolay Aleksandrov
>>>> <nikolay@cumulusnetworks.com>; brouer@redhat.com
>>>> Subject: Re: [PATCH net-next RFC WIP] Patch for XDP support for
>>> virtio_net
>>>>
>>>> On Sat, 29 Oct 2016 13:25:14 +0200
>>>> Thomas Graf <tgraf@suug.ch> wrote:
>>>>
>>>>> On 10/28/16 at 08:51pm, Shrijeet Mukherjee wrote:
>>>>>> Generally agree, but SRIOV nics with multiple queues can end up in a
>>>>>> bad spot if each buffer was 4K right ? I see a specific page pool to
>>>>>> be used by queues which are enabled for XDP as the easiest to swing
>>>>>> solution that way the memory overhead can be restricted to enabled
>>>>>> queues and shared access issues can be restricted to skb's using
>>> that
>>>> pool no ?
>>>>
>>>> Yes, that is why that I've been arguing so strongly for having the
>>> flexibility to
>>>> attach a XDP program per RX queue, as this only change the memory model
>>>> for this one queue.
>>>>
>>>>
>>>>> Isn't this clearly a must anyway? I may be missing something
>>>>> fundamental here so please enlighten me :-)
>>>>>
>>>>> If we dedicate a page per packet, that could translate to 14M*4K worth
>>>>> of memory being mapped per second for just a 10G NIC under DoS attack.
>>>>> How can one protect such as system? Is the assumption that we can
>>>>> always drop such packets quickly enough before we start dropping
>>>>> randomly due to memory pressure? If a handshake is required to
>>>>> determine validity of a packet then that is going to be difficult.
>>>>
>>>> Under DoS attacks you don't run out of memory, because a diverse set of
>>>> socket memory limits/accounting avoids that situation. What does happen
>>>> is the maximum achievable PPS rate is directly dependent on the
>>>> time you spend on each packet. This use of CPU resources (and
>>>> hitting mem-limits-safe-guards) push-back on the drivers speed to
>>> process
>>>> the RX ring. In effect, packets are dropped in the NIC HW as RX-ring
>>> queue
>>>> is not emptied fast-enough.
>>>>
>>>> Given you don't control what HW drops, the attacker will "successfully"
>>>> cause your good traffic to be among the dropped packets.
>>>>
>>>> This is where XDP change the picture. If you can express (by eBPF) a
>>> filter
>>>> that can separate "bad" vs "good" traffic, then you can take back
>>> control.
>>>> Almost like controlling what traffic the HW should drop.
>>>> Given the cost of XDP-eBPF filter + serving regular traffic does not use
>>> all of
>>>> your CPU resources, you have overcome the attack.
>>>>
>>>> --
>>> Jesper, John et al .. to make this a little concrete I am going to spin
>>> up a v2 which has only bigbuffers mode enabled for xdp acceleration, all
>>> other modes will reject the xdp ndo ..
>>>
>>> Do we have agreement on that model ?
>>>
>>> It will need that all vhost implementations will need to start with
>>> mergeable buffers disabled to get xdp goodness, but that sounds like a
>>> safe thing to do for now ..
>>
>> It's ok for experimentation, but really after speaking with Alexei it's
>> clear to me that xdp should have a separate code path in the driver,
>> e.g. the separation between modes is something that does not
>> make sense for xdp.
>>
>> The way I imagine it working:
>
> OK I tried to make some sense out of this and get it working,
>
>>
>> - when XDP is attached disable all LRO using VIRTIO_NET_CTRL_GUEST_OFFLOADS_SET
>> (not used by driver so far, designed to allow dynamic LRO control with
>> ethtool)
>
> I see there is a UAPI bit for this but I guess we also need to add
> support to vhost as well? Seems otherwise we may just drop a bunch
> of packets on the floor out of handle_rx() when recvmsg returns larger
> than a page size. Or did I read this wrong...
>
>> - start adding page-sized buffers
>
> I started to mangle add_recvbuf_big() and receive_big() here and this
> didn't seem too bad.
>
>> - do something with non-page-sized buffers added previously - what
>> exactly? copy I guess? What about LRO packets that are too large -
>> can we drop or can we split them up?
>
> hmm not sure I understand this here. With LRO disabled and mergeable
> buffers disabled all packets should fit in a page correct?
>
> With LRO enabled case I guess to start with we block XDP from being
> loaded for the same reason we don't allow jumbo frames on physical
> nics.
>
>>
>> I'm fine with disabling XDP for some configurations as the first step,
>> and we can add that support later.
>>
>
> In order for this to work though I guess we need to be able to
> dynamically disable mergeable buffers at the moment I just commented
> it out of the features list and fixed up virtio_has_features so it
> wont bug_on.
>
>> Ideas about mergeable buffers (optional):
>>
>> At the moment mergeable buffers can't be disabled dynamically.
>> They do bring a small benefit for XDP if host MTU is large (see below)
>> and aren't hard to support:
>> - if header is by itself skip 1st page
>> - otherwise copy all data into first page
>> and it's nicer not to add random limitations that require guest reboot.
>> It might make sense to add a command that disables/enabled
>> mergeable buffers dynamically but that's for newer hosts.
>
> Yep it seems disabling mergeable buffers solves this but didn't look at
> it too closely. I'll look closer tomorrow.
>
Actually after some more research I can't see why with LRO off using
mergeable buffers causes any issues? For XDP all we need is a pointer
to contiguous data and the length. Seems all conditions are met here
assuming I'm reading the code correctly.
Any more hints on this comment "otherwise copy all data into first
page" is required?
>>
>> Spec does not require it but in practice most hosts put all data
>> in the 1st page or all in the 2nd page so the copy will be nop
>> for these cases.
>>
>> Large host MTU - newer hosts report the host MTU, older ones don't.
>> Using mergeable buffers we can at least detect this case
>> (and then what? drop I guess).
>>
>
> The physical nics just refuse to load XDP with large MTU. Any reason
> not to negotiate the mtu with the guest so that the guest can force
> this?
>
>>
>>
>>
>
> Thanks,
> John
>
^ permalink raw reply
* [PATCH net-next] sock: do not set sk_err in sock_dequeue_err_skb
From: Soheil Hassas Yeganeh @ 2016-11-03 22:24 UTC (permalink / raw)
To: davem, netdev; +Cc: edumazet, willemb, ncardwell, Soheil Hassas Yeganeh
From: Soheil Hassas Yeganeh <soheil@google.com>
Do not set sk_err when dequeuing errors from the error queue.
Doing so results in:
a) Bugs: By overwriting existing sk_err values, it possibly
hides legitimate errors. It is also incorrect when local
errors are queued with ip_local_error. That happens in the
context of a system call, which already returns the error
code.
b) Inconsistent behavior: When there are pending errors on
the error queue, sk_err is sometimes 0 (e.g., for
the first timestamp on the error queue) and sometimes
set to an error code (after dequeuing the first
timestamp).
c) Suboptimality: Setting sk_err to ENOMSG on simple
TX timestamps can abort parallel reads and writes.
Removing this line doesn't break userspace. This is because
userspace code cannot rely on sk_err for detecting whether
there is something on the error queue. Except for ICMP messages
received for UDP and RAW, sk_err is not set at enqueue time,
and as a result sk_err can be 0 while there are plenty of
errors on the error queue.
For ICMP packets in UDP and RAW, sk_err is set when they are
enqueued on the error queue, but that does not result in aborting
reads and writes. For such cases, sk_err is only readable via
getsockopt(SO_ERROR) which will reset the value of sk_err on
its own. More importantly, prior to this patch,
recvmsg(MSG_ERRQUEUE) has a race on setting sk_err (i.e.,
sk_err is set by sock_dequeue_err_skb without atomic ops or
locks) which can store 0 in sk_err even when we have ICMP
messages pending. Removing this line from sock_dequeue_err_skb
eliminates that race.
Signed-off-by: Soheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: Neal Cardwell <ncardwell@google.com>
---
net/core/skbuff.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 1e3e008..0b2a6e9 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -3725,7 +3725,6 @@ struct sk_buff *sock_dequeue_err_skb(struct sock *sk)
err = SKB_EXT_ERR(skb_next)->ee.ee_errno;
spin_unlock_irqrestore(&q->lock, flags);
- sk->sk_err = err;
if (err)
sk->sk_error_report(sk);
--
2.8.0.rc3.226.g39d4020
^ permalink raw reply related
* Re: [PATCH for-next V2 00/15][PULL request] Mellanox mlx5 core driver updates 2016-10-25
From: Doug Ledford @ 2016-11-03 22:33 UTC (permalink / raw)
To: David Miller, saeedm@mellanox.com
Cc: netdev@vger.kernel.org, linux-rdma@vger.kernel.org,
ogerlitz@mellanox.com, leonro@mellanox.com, talal@mellanox.com,
matanb@mellanox.com
In-Reply-To: <20161030.173522.1563055187413070239.davem@davemloft.net>
[-- Attachment #1.1: Type: text/plain, Size: 637 bytes --]
On 10/30/16 3:35 PM, David Miller wrote:
> From: Saeed Mahameed <saeedm@mellanox.com>
> Date: Sun, 30 Oct 2016 23:21:53 +0200
>
>> This series contains some updates and fixes of mlx5 core and
>> IB drivers with the addition of two features that demand
>> new low level commands and infrastructure updates.
>> - SRIOV VF max rate limit support
>> - mlx5e tc support for FWD rules with counter.
>>
>> Needed for both net and rdma subsystems.
>
> Pulled, thanks.
Thanks, done here as well.
--
Doug Ledford <dledford@redhat.com> GPG Key ID: 0E572FDD
Red Hat, Inc.
100 E. Davie St
Raleigh, NC 27601 USA
[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 907 bytes --]
^ permalink raw reply
* Re: [PATCH net-next RFC WIP] Patch for XDP support for virtio_net
From: Michael S. Tsirkin @ 2016-11-03 22:42 UTC (permalink / raw)
To: John Fastabend
Cc: Shrijeet Mukherjee, Jesper Dangaard Brouer, Thomas Graf,
Alexei Starovoitov, Jakub Kicinski, David Miller, alexander.duyck,
shrijeet, tom, netdev, Roopa Prabhu, Nikolay Aleksandrov, aconole
In-Reply-To: <581ADCD1.3040404@gmail.com>
On Wed, Nov 02, 2016 at 11:44:33PM -0700, John Fastabend wrote:
> On 16-11-02 09:11 PM, Michael S. Tsirkin wrote:
> > On Wed, Nov 02, 2016 at 06:28:34PM -0700, Shrijeet Mukherjee wrote:
> >>> -----Original Message-----
> >>> From: Jesper Dangaard Brouer [mailto:brouer@redhat.com]
> >>> Sent: Wednesday, November 2, 2016 7:27 AM
> >>> To: Thomas Graf <tgraf@suug.ch>
> >>> Cc: Shrijeet Mukherjee <shm@cumulusnetworks.com>; Alexei Starovoitov
> >>> <alexei.starovoitov@gmail.com>; Jakub Kicinski <kubakici@wp.pl>; John
> >>> Fastabend <john.fastabend@gmail.com>; David Miller
> >>> <davem@davemloft.net>; alexander.duyck@gmail.com; mst@redhat.com;
> >>> shrijeet@gmail.com; tom@herbertland.com; netdev@vger.kernel.org;
> >>> Roopa Prabhu <roopa@cumulusnetworks.com>; Nikolay Aleksandrov
> >>> <nikolay@cumulusnetworks.com>; brouer@redhat.com
> >>> Subject: Re: [PATCH net-next RFC WIP] Patch for XDP support for
> >> virtio_net
> >>>
> >>> On Sat, 29 Oct 2016 13:25:14 +0200
> >>> Thomas Graf <tgraf@suug.ch> wrote:
> >>>
> >>>> On 10/28/16 at 08:51pm, Shrijeet Mukherjee wrote:
> >>>>> Generally agree, but SRIOV nics with multiple queues can end up in a
> >>>>> bad spot if each buffer was 4K right ? I see a specific page pool to
> >>>>> be used by queues which are enabled for XDP as the easiest to swing
> >>>>> solution that way the memory overhead can be restricted to enabled
> >>>>> queues and shared access issues can be restricted to skb's using
> >> that
> >>> pool no ?
> >>>
> >>> Yes, that is why that I've been arguing so strongly for having the
> >> flexibility to
> >>> attach a XDP program per RX queue, as this only change the memory model
> >>> for this one queue.
> >>>
> >>>
> >>>> Isn't this clearly a must anyway? I may be missing something
> >>>> fundamental here so please enlighten me :-)
> >>>>
> >>>> If we dedicate a page per packet, that could translate to 14M*4K worth
> >>>> of memory being mapped per second for just a 10G NIC under DoS attack.
> >>>> How can one protect such as system? Is the assumption that we can
> >>>> always drop such packets quickly enough before we start dropping
> >>>> randomly due to memory pressure? If a handshake is required to
> >>>> determine validity of a packet then that is going to be difficult.
> >>>
> >>> Under DoS attacks you don't run out of memory, because a diverse set of
> >>> socket memory limits/accounting avoids that situation. What does happen
> >>> is the maximum achievable PPS rate is directly dependent on the
> >>> time you spend on each packet. This use of CPU resources (and
> >>> hitting mem-limits-safe-guards) push-back on the drivers speed to
> >> process
> >>> the RX ring. In effect, packets are dropped in the NIC HW as RX-ring
> >> queue
> >>> is not emptied fast-enough.
> >>>
> >>> Given you don't control what HW drops, the attacker will "successfully"
> >>> cause your good traffic to be among the dropped packets.
> >>>
> >>> This is where XDP change the picture. If you can express (by eBPF) a
> >> filter
> >>> that can separate "bad" vs "good" traffic, then you can take back
> >> control.
> >>> Almost like controlling what traffic the HW should drop.
> >>> Given the cost of XDP-eBPF filter + serving regular traffic does not use
> >> all of
> >>> your CPU resources, you have overcome the attack.
> >>>
> >>> --
> >> Jesper, John et al .. to make this a little concrete I am going to spin
> >> up a v2 which has only bigbuffers mode enabled for xdp acceleration, all
> >> other modes will reject the xdp ndo ..
> >>
> >> Do we have agreement on that model ?
> >>
> >> It will need that all vhost implementations will need to start with
> >> mergeable buffers disabled to get xdp goodness, but that sounds like a
> >> safe thing to do for now ..
> >
> > It's ok for experimentation, but really after speaking with Alexei it's
> > clear to me that xdp should have a separate code path in the driver,
> > e.g. the separation between modes is something that does not
> > make sense for xdp.
> >
> > The way I imagine it working:
>
> OK I tried to make some sense out of this and get it working,
>
> >
> > - when XDP is attached disable all LRO using VIRTIO_NET_CTRL_GUEST_OFFLOADS_SET
> > (not used by driver so far, designed to allow dynamic LRO control with
> > ethtool)
>
> I see there is a UAPI bit for this but I guess we also need to add
> support to vhost as well? Seems otherwise we may just drop a bunch
> of packets on the floor out of handle_rx() when recvmsg returns larger
> than a page size. Or did I read this wrong...
It's already supported host side. However you might
get some packets that were in flight when you attached.
> > - start adding page-sized buffers
>
> I started to mangle add_recvbuf_big() and receive_big() here and this
> didn't seem too bad.
I imagine it won't be ATM but I think we'll need to support
mrg buffers with time and then it will be messy.
Besides, it's not an architectural thing that receive_big
uses page sized buffers, it could use any size.
So a separate path just for xdp would be better imho.
> > - do something with non-page-sized buffers added previously - what
> > exactly? copy I guess? What about LRO packets that are too large -
> > can we drop or can we split them up?
>
> hmm not sure I understand this here. With LRO disabled and mergeable
> buffers disabled all packets should fit in a page correct?
Assuing F_MTU is negotiated and MTU field is small enough, yes.
But if you disable mrg buffers dynamically you will get some packets
in buffers that were added before the disable.
Similarly for disabling LRO dynamically.
> With LRO enabled case I guess to start with we block XDP from being
> loaded for the same reason we don't allow jumbo frames on physical
> nics.
If you ask that host disables the capability, then yes, it's easy.
Let's do that for now, it's a start.
> >
> > I'm fine with disabling XDP for some configurations as the first step,
> > and we can add that support later.
> >
>
> In order for this to work though I guess we need to be able to
> dynamically disable mergeable buffers at the moment I just commented
> it out of the features list and fixed up virtio_has_features so it
> wont bug_on.
For now we can just set mrg_rxbuf=off on qemu command line, and
fail XDP attach if not there. I think we'll be able to support it
long term but you will need host side changes, or fully reset
device and reconfigure it.
> > Ideas about mergeable buffers (optional):
> >
> > At the moment mergeable buffers can't be disabled dynamically.
> > They do bring a small benefit for XDP if host MTU is large (see below)
> > and aren't hard to support:
> > - if header is by itself skip 1st page
> > - otherwise copy all data into first page
> > and it's nicer not to add random limitations that require guest reboot.
> > It might make sense to add a command that disables/enabled
> > mergeable buffers dynamically but that's for newer hosts.
>
> Yep it seems disabling mergeable buffers solves this but didn't look at
> it too closely. I'll look closer tomorrow.
>
> >
> > Spec does not require it but in practice most hosts put all data
> > in the 1st page or all in the 2nd page so the copy will be nop
> > for these cases.
> >
> > Large host MTU - newer hosts report the host MTU, older ones don't.
> > Using mergeable buffers we can at least detect this case
> > (and then what? drop I guess).
> >
>
> The physical nics just refuse to load XDP with large MTU.
So let's do the same for now, unfortunately you don't know
the MTU unless _F_MTU is negitiated and QEMU does not
implement that yet, but it's easy to add.
In fact I suspect Aaron (cc) has an implementation since
he posted a patch implementing that.
Aaron could you post it pls?
> Any reason
> not to negotiate the mtu with the guest so that the guest can force
> this?
There are generally many guests and many NICs on the host.
A big packet arrives, what do you want to do with it?
We probably want to build propagating MTU across all VMs and NICs
but let's get a basic thing merged first.
> >
> >
> >
>
> Thanks,
> John
^ permalink raw reply
* [PATCH net] bpf: fix htab map destruction when extra reserve is in use
From: Daniel Borkmann @ 2016-11-03 23:01 UTC (permalink / raw)
To: davem; +Cc: netdev, ast, dvyukov, Daniel Borkmann
Commit a6ed3ea65d98 ("bpf: restore behavior of bpf_map_update_elem")
added an extra per-cpu reserve to the hash table map to restore old
behaviour from pre prealloc times. When non-prealloc is in use for a
map, then problem is that once a hash table extra element has been
linked into the hash-table, and the hash table is destroyed due to
refcount dropping to zero, then htab_map_free() -> delete_all_elements()
will walk the whole hash table and drop all elements via htab_elem_free().
The problem is that the element from the extra reserve is first fed
to the wrong backend allocator and eventually freed twice.
Fixes: a6ed3ea65d98 ("bpf: restore behavior of bpf_map_update_elem")
Reported-by: Dmitry Vyukov <dvyukov@google.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Alexei Starovoitov <ast@kernel.org>
---
kernel/bpf/hashtab.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/kernel/bpf/hashtab.c b/kernel/bpf/hashtab.c
index 570eeca..ad1bc67 100644
--- a/kernel/bpf/hashtab.c
+++ b/kernel/bpf/hashtab.c
@@ -687,7 +687,8 @@ static void delete_all_elements(struct bpf_htab *htab)
hlist_for_each_entry_safe(l, n, head, hash_node) {
hlist_del_rcu(&l->hash_node);
- htab_elem_free(htab, l);
+ if (l->state != HTAB_EXTRA_ELEM_USED)
+ htab_elem_free(htab, l);
}
}
}
--
1.9.3
^ permalink raw reply related
* net/netlink: another global-out-of-bounds in genl_family_rcv_msg/validate_nla
From: Andrey Konovalov @ 2016-11-03 23:04 UTC (permalink / raw)
To: David S. Miller, Nicolas Dichtel, LKML, Matti Vaittinen,
Tycho Andersen, stephen hemminger, Tom Herbert, Florian Westphal,
netdev
Cc: Dmitry Vyukov, Alexander Potapenko, Kostya Serebryany,
Eric Dumazet, syzkaller
Hi,
I've got the following error report while running the syzkaller fuzzer:
BUG: KASAN: global-out-of-bounds in validate_nla+0x49b/0x4e0 at addr
ffffffff84452de0
Read of size 2 by task syz-executor/19055
Address belongs to variable ip_vs_cmd_policy+0x20/0x40
CPU: 1 PID: 19055 Comm: syz-executor Not tainted 4.9.0-rc3+ #350
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS Bochs 01/01/2011
ffff88006b547638 ffffffff81b46934 ffff88006b5476c8 ffffffff847a361f
ffffffff84452dc0 ffffffff84452de0 ffff88006b5476b8 ffffffff8150ac7c
ffffffff859bdf80 ffffffff85f44280 ffff88003df282c0 0000000000000292
Call Trace:
[< inline >] __dump_stack lib/dump_stack.c:15
[<ffffffff81b46934>] dump_stack+0xb3/0x10f lib/dump_stack.c:51
[< inline >] print_address_description mm/kasan/report.c:204
[<ffffffff8150ac7c>] kasan_report_error+0x49c/0x4d0 mm/kasan/report.c:283
[< inline >] kasan_report mm/kasan/report.c:303
[<ffffffff8150ad2e>] __asan_report_load2_noabort+0x3e/0x40
mm/kasan/report.c:322
[<ffffffff81be27eb>] validate_nla+0x49b/0x4e0 lib/nlattr.c:41
[<ffffffff81be2ab5>] nla_parse+0x115/0x280 lib/nlattr.c:195
[< inline >] nlmsg_parse include/net/netlink.h:386
[<ffffffff82dc2723>] genl_family_rcv_msg+0x543/0xc80
net/netlink/genetlink.c:613
[<ffffffff82dc3016>] genl_rcv_msg+0x1b6/0x270 net/netlink/genetlink.c:658
[<ffffffff82dc10a0>] netlink_rcv_skb+0x2c0/0x3b0 net/netlink/af_netlink.c:2281
[<ffffffff82dc21c8>] genl_rcv+0x28/0x40 net/netlink/genetlink.c:669
[< inline >] netlink_unicast_kernel net/netlink/af_netlink.c:1214
[<ffffffff82dbf959>] netlink_unicast+0x5a9/0x880 net/netlink/af_netlink.c:1240
[<ffffffff82dc05e7>] netlink_sendmsg+0x9b7/0xce0 net/netlink/af_netlink.c:1786
[< inline >] sock_sendmsg_nosec net/socket.c:606
[<ffffffff82b6f75c>] sock_sendmsg+0xcc/0x110 net/socket.c:616
[<ffffffff82b6f9c1>] sock_write_iter+0x221/0x3b0 net/socket.c:814
[< inline >] new_sync_write fs/read_write.c:499
[<ffffffff8151bd44>] __vfs_write+0x334/0x570 fs/read_write.c:512
[<ffffffff8151f85b>] vfs_write+0x17b/0x500 fs/read_write.c:560
[< inline >] SYSC_write fs/read_write.c:607
[<ffffffff81523184>] SyS_write+0xd4/0x1a0 fs/read_write.c:599
[<ffffffff81006465>] do_syscall_64+0x195/0x490 arch/x86/entry/common.c:280
[<ffffffff83fc0409>] entry_SYSCALL64_slow_path+0x25/0x25
Memory state around the buggy address:
ffffffff84452c80: fa fa fa fa 00 00 00 00 00 00 04 fa fa fa fa fa
ffffffff84452d00: 00 00 00 00 00 00 04 fa fa fa fa fa 00 00 00 00
>ffffffff84452d80: 04 fa fa fa fa fa fa fa 00 00 00 04 fa fa fa fa
^
ffffffff84452e00: 00 fa fa fa fa fa fa fa 00 00 fa fa fa fa fa fa
ffffffff84452e80: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
==================================================================
This time the out-of-bounds is on the ip_vs_cmd_policy variable.
On commit 0c183d92b20b5c84ca655b45ef57b3318b83eb9e (Oct 31).
Thanks!
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox