* [PATCH v3] lib: fix multiple strlcpy definition
From: Baruch Siach @ 2017-10-09 5:49 UTC (permalink / raw)
To: Stephen Hemminger; +Cc: netdev, Phil Sutter, Baruch Siach
Some C libraries, like uClibc and musl, provide BSD compatible
strlcpy(). Add check_strlcpy() to configure, and avoid defining strlcpy
and strlcat when the C library provides them.
This fixes the following static link error with uClibc-ng:
.../sysroot/usr/lib/libc.a(strlcpy.os): In function `strlcpy':
strlcpy.c:(.text+0x0): multiple definition of `strlcpy'
../lib/libutil.a(utils.o):utils.c:(.text+0x1ddc): first defined here
collect2: error: ld returned 1 exit status
Acked-by: Phil Sutter <phil@nwl.cc>
Signed-off-by: Baruch Siach <baruch@tkos.co.il>
---
v3: Set CFLAGS directly in config.mk
v2: Fix the order of strlcpy parameters
---
configure | 24 ++++++++++++++++++++++++
lib/utils.c | 2 ++
2 files changed, 26 insertions(+)
diff --git a/configure b/configure
index 7be8fb113cc9..f0668ab3f7e9 100755
--- a/configure
+++ b/configure
@@ -326,6 +326,27 @@ EOF
rm -f $TMPDIR/dbtest.c $TMPDIR/dbtest
}
+check_strlcpy()
+{
+ cat >$TMPDIR/strtest.c <<EOF
+#include <string.h>
+int main(int argc, char **argv) {
+ char dst[10];
+ strlcpy(dst, "test", sizeof(dst));
+ return 0;
+}
+EOF
+ $CC -I$INCLUDE -o $TMPDIR/strtest $TMPDIR/strtest.c >/dev/null 2>&1
+ if [ $? -eq 0 ]
+ then
+ echo "no"
+ else
+ echo 'CFLAGS += -DNEED_STRLCPY' >>$CONFIG
+ echo "yes"
+ fi
+ rm -f $TMPDIR/strtest.c $TMPDIR/strtest
+}
+
quiet_config()
{
cat <<EOF
@@ -397,6 +418,9 @@ check_mnl
echo -n "Berkeley DB: "
check_berkeley_db
+echo -n "need for strlcpy: "
+check_strlcpy
+
echo
echo -n "docs:"
check_docs
diff --git a/lib/utils.c b/lib/utils.c
index 0cf99619c302..632fd0ddac82 100644
--- a/lib/utils.c
+++ b/lib/utils.c
@@ -1260,6 +1260,7 @@ int get_real_family(int rtm_type, int rtm_family)
return rtm_family;
}
+#ifdef NEED_STRLCPY
size_t strlcpy(char *dst, const char *src, size_t size)
{
size_t srclen = strlen(src);
@@ -1282,3 +1283,4 @@ size_t strlcat(char *dst, const char *src, size_t size)
return dlen + strlcpy(dst + dlen, src, size - dlen);
}
+#endif
--
2.14.2
^ permalink raw reply related
* IRREVOCABLE PAYMENT ORDER.pdf
From: Dr. Owen Douglas @ 2017-10-09 0:12 UTC (permalink / raw)
In-Reply-To: <34be7bb2e296579de8756eb7e5261ca2@incoming.mhcable.com>
[-- Attachment #1: IRREVOCABLE PAYMENT ORDER.pdf --]
[-- Type: application/pdf, Size: 181455 bytes --]
^ permalink raw reply
* Re: [PATCH net-next 0/2] net: defer cgroups init to accept()
From: Eric Dumazet @ 2017-10-09 4:47 UTC (permalink / raw)
To: David S . Miller
Cc: netdev, Eric Dumazet, Eric Dumazet, Johannes Weiner, Tejun Heo,
John Sperbeck
In-Reply-To: <20171009044452.20564-1-edumazet@google.com>
On Sun, Oct 8, 2017 at 9:44 PM, Eric Dumazet <edumazet@google.com> wrote:
> After TCP 3WHS became lockless, we should not attempt cgroup games
> from sk_clone_lock() since listener/cgroup might be already gone.
>
> Move this business to inet_csk_accept() where we have
> the guarantee both parent and child exist.
>
> Many thanks to John Sperbeck for spotting these issues
>
> Eric Dumazet (2):
> net: memcontrol: defer call to mem_cgroup_sk_alloc()
> net: defer call to cgroup_sk_alloc()
This was based on net tree, but I used the wrong script, and thus this
has the [PATCH net-next] tag.
Sorry for the confusion, but I guess this also can be applied to
net-next since this is not a recent regression.
^ permalink raw reply
* Re: netlink backwards compatibility in userspace tools
From: David Miller @ 2017-10-09 4:47 UTC (permalink / raw)
To: Jason; +Cc: netdev, linux-kernel, dkg
In-Reply-To: <CAHmME9oixZtPVdH24KJQ9NaTuf_ECAOoHwQhuA+Fy-BX+F_3dw@mail.gmail.com>
From: "Jason A. Donenfeld" <Jason@zx2c4.com>
Date: Fri, 29 Sep 2017 12:22:42 +0200
> One handy aspect of Netlink is that it's backwards compatible. This
> means that you can run old userspace utilities on new kernels, even if
> the new kernel supports new features and netlink attributes. The wire
> format is stable enough that the data marshaled can be extended
> without breaking compat. Neat.
>
> I was wondering, though, what you think the best stance is toward
> these old userspace utilities. What should they do if the kernel sends
> it netlink attributes that it does not recognize? At the moment, I'm
> doing something like this:
>
> static void warn_unrecognized(void)
> {
> static bool once = false;
> if (once)
> return;
> once = true;
> fprintf(stderr,
> "Warning: this program received from your kernel one or more\n"
> "attributes that it did not recognize. It is possible that\n"
> "this version of wg(8) is older than your kernel. You may\n"
> "want to update this program.\n");
> }
>
> This seems like a somewhat sensible warning, but then I wonder about
> distributions like Debian, which has a long stable life cycle, so it
> frequently has very old tools (ancient iproute2 for example). Then,
> VPS providers have these Debian images run on top of newer kernels.
> People in this situation would undoubtedly see the above warning a lot
> and not be able to do anything about it. Not horrible, but a bit
> annoying. Is this an okay annoyance? Or is it advised to just have no
> warning at all? One idea would be to put it behind an environment
> variable flag, but I don't like too many nobs.
>
> I'm generally wondering about attitudes toward this kind of userspace
> program behavior in response to newer kernels.
Generally, yes you should simply ignore attributes you don't understand.
But we keep coming back to this issue, because it's not always the best
thing to do.
For example, let's say you have settings X and Y for object A.
User A has a newer tool and is able to set both X and Y, as well as
see them in dumps. And let's further assume that Y's setting has some
kind of influence on the behavior of X.
User B has an older tool, and sees X but not Y because Y is not
understood by the older tool. User B will not be able to figure out
why X is not behaving the way they expect it to, because of the loss
of information.
Similar, even more serious, issues arise when setting values. User B
can set X and wonder why it's not doing what they expect it to do
because of setting Y which they can't even see with their tools.
For this reason it might be beneficical to at least say to the user
"Warning, I've seen one or more unrecognized netlink attributes."
so that there is at least a chance for the user to figure out what
might be happening to them.
^ permalink raw reply
* Re: [PATCHv2 net-next] openvswitch: Add erspan tunnel support.
From: Pravin Shelar @ 2017-10-09 4:45 UTC (permalink / raw)
To: William Tu; +Cc: Linux Kernel Network Developers
In-Reply-To: <1507161792-18340-1-git-send-email-u9012063@gmail.com>
On Wed, Oct 4, 2017 at 5:03 PM, William Tu <u9012063@gmail.com> wrote:
> Add erspan netlink interface for OVS.
>
> Signed-off-by: William Tu <u9012063@gmail.com>
> Cc: Pravin B Shelar <pshelar@ovn.org>
> ---
> v1->v2: remove unnecessary compat code.
Looks good to me.
^ permalink raw reply
* [PATCH net-next 2/2] net: defer call to cgroup_sk_alloc()
From: Eric Dumazet @ 2017-10-09 4:44 UTC (permalink / raw)
To: David S . Miller
Cc: netdev, Eric Dumazet, Eric Dumazet, Johannes Weiner, Tejun Heo,
John Sperbeck
In-Reply-To: <20171009044452.20564-1-edumazet@google.com>
sk_clone_lock() might run while TCP/DCCP listener already vanished.
In order to prevent use after free, it is better to defer cgroup_sk_alloc()
to the point we know both parent and child exist, and from process context.
Fixes: e994b2f0fb92 ("tcp: do not lock listener to process SYN packets")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Tejun Heo <tj@kernel.org>
---
kernel/cgroup/cgroup.c | 11 -----------
net/core/sock.c | 3 +--
net/ipv4/inet_connection_sock.c | 5 +++++
3 files changed, 6 insertions(+), 13 deletions(-)
diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c
index 44857278eb8aa6a2bbf27b7eb12137ef42628170..3380a3e49af501e457991b2823020494cf32af80 100644
--- a/kernel/cgroup/cgroup.c
+++ b/kernel/cgroup/cgroup.c
@@ -5709,17 +5709,6 @@ void cgroup_sk_alloc(struct sock_cgroup_data *skcd)
if (cgroup_sk_alloc_disabled)
return;
- /* Socket clone path */
- if (skcd->val) {
- /*
- * We might be cloning a socket which is left in an empty
- * cgroup and the cgroup might have already been rmdir'd.
- * Don't use cgroup_get_live().
- */
- cgroup_get(sock_cgroup_ptr(skcd));
- return;
- }
-
rcu_read_lock();
while (true) {
diff --git a/net/core/sock.c b/net/core/sock.c
index 70c6ccbdf49f2f8a5a0f7c41c7849ea01459be50..4499e31538132ed59a16d92e6f6b923e776df84e 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -1680,6 +1680,7 @@ struct sock *sk_clone_lock(const struct sock *sk, const gfp_t priority)
/* sk->sk_memcg will be populated at accept() time */
newsk->sk_memcg = NULL;
+ memset(&newsk->sk_cgrp_data, 0, sizeof(newsk->sk_cgrp_data));
atomic_set(&newsk->sk_drops, 0);
newsk->sk_send_head = NULL;
@@ -1718,8 +1719,6 @@ struct sock *sk_clone_lock(const struct sock *sk, const gfp_t priority)
newsk->sk_incoming_cpu = raw_smp_processor_id();
atomic64_set(&newsk->sk_cookie, 0);
- cgroup_sk_alloc(&newsk->sk_cgrp_data);
-
/*
* Before updating sk_refcnt, we must commit prior changes to memory
* (Documentation/RCU/rculist_nulls.txt for details)
diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
index 67aec7a106860b26c929fea1624d652c87972f04..d32c74507314cc4b91d040de8e877e4bd8204106 100644
--- a/net/ipv4/inet_connection_sock.c
+++ b/net/ipv4/inet_connection_sock.c
@@ -26,6 +26,8 @@
#include <net/tcp.h>
#include <net/sock_reuseport.h>
#include <net/addrconf.h>
+#include <net/cls_cgroup.h>
+#include <net/netprio_cgroup.h>
#ifdef INET_CSK_DEBUG
const char inet_csk_timer_bug_msg[] = "inet_csk BUG: unknown timer value\n";
@@ -476,6 +478,9 @@ struct sock *inet_csk_accept(struct sock *sk, int flags, int *err, bool kern)
spin_unlock_bh(&queue->fastopenq.lock);
}
mem_cgroup_sk_alloc(newsk);
+ cgroup_sk_alloc(&newsk->sk_cgrp_data);
+ sock_update_classid(&newsk->sk_cgrp_data);
+ sock_update_netprioidx(&newsk->sk_cgrp_data);
out:
release_sock(sk);
if (req)
--
2.14.2.920.gcf0c67979c-goog
^ permalink raw reply related
* [PATCH net-next 1/2] net: memcontrol: defer call to mem_cgroup_sk_alloc()
From: Eric Dumazet @ 2017-10-09 4:44 UTC (permalink / raw)
To: David S . Miller
Cc: netdev, Eric Dumazet, Eric Dumazet, Johannes Weiner, Tejun Heo,
John Sperbeck
In-Reply-To: <20171009044452.20564-1-edumazet@google.com>
Instead of calling mem_cgroup_sk_alloc() from BH context,
it is better to call it from inet_csk_accept() in process context.
Not only this removes code in mem_cgroup_sk_alloc(), but it also
fixes a bug since listener might have been dismantled and css_get()
might cause a use-after-free.
Fixes: e994b2f0fb92 ("tcp: do not lock listener to process SYN packets")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Tejun Heo <tj@kernel.org>
---
mm/memcontrol.c | 15 ---------------
net/core/sock.c | 5 ++++-
net/ipv4/inet_connection_sock.c | 1 +
3 files changed, 5 insertions(+), 16 deletions(-)
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index d5f3a62887cf958f6b657c0f542f0cf2c3e86e8d..661f046ad3181f65eccfd9bf3832e395e27aa226 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -5828,21 +5828,6 @@ void mem_cgroup_sk_alloc(struct sock *sk)
if (!mem_cgroup_sockets_enabled)
return;
- /*
- * Socket cloning can throw us here with sk_memcg already
- * filled. It won't however, necessarily happen from
- * process context. So the test for root memcg given
- * the current task's memcg won't help us in this case.
- *
- * Respecting the original socket's memcg is a better
- * decision in this case.
- */
- if (sk->sk_memcg) {
- BUG_ON(mem_cgroup_is_root(sk->sk_memcg));
- css_get(&sk->sk_memcg->css);
- return;
- }
-
rcu_read_lock();
memcg = mem_cgroup_from_task(current);
if (memcg == root_mem_cgroup)
diff --git a/net/core/sock.c b/net/core/sock.c
index 23953b741a41fbcf4a6ffb0dd5bf05bd5266b99d..70c6ccbdf49f2f8a5a0f7c41c7849ea01459be50 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -1677,6 +1677,10 @@ struct sock *sk_clone_lock(const struct sock *sk, const gfp_t priority)
newsk->sk_dst_pending_confirm = 0;
newsk->sk_wmem_queued = 0;
newsk->sk_forward_alloc = 0;
+
+ /* sk->sk_memcg will be populated at accept() time */
+ newsk->sk_memcg = NULL;
+
atomic_set(&newsk->sk_drops, 0);
newsk->sk_send_head = NULL;
newsk->sk_userlocks = sk->sk_userlocks & ~SOCK_BINDPORT_LOCK;
@@ -1714,7 +1718,6 @@ struct sock *sk_clone_lock(const struct sock *sk, const gfp_t priority)
newsk->sk_incoming_cpu = raw_smp_processor_id();
atomic64_set(&newsk->sk_cookie, 0);
- mem_cgroup_sk_alloc(newsk);
cgroup_sk_alloc(&newsk->sk_cgrp_data);
/*
diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
index c039c937ba90c7aec39ba2687bceb8253ead70aa..67aec7a106860b26c929fea1624d652c87972f04 100644
--- a/net/ipv4/inet_connection_sock.c
+++ b/net/ipv4/inet_connection_sock.c
@@ -475,6 +475,7 @@ struct sock *inet_csk_accept(struct sock *sk, int flags, int *err, bool kern)
}
spin_unlock_bh(&queue->fastopenq.lock);
}
+ mem_cgroup_sk_alloc(newsk);
out:
release_sock(sk);
if (req)
--
2.14.2.920.gcf0c67979c-goog
^ permalink raw reply related
* [PATCH net-next 0/2] net: defer cgroups init to accept()
From: Eric Dumazet @ 2017-10-09 4:44 UTC (permalink / raw)
To: David S . Miller
Cc: netdev, Eric Dumazet, Eric Dumazet, Johannes Weiner, Tejun Heo,
John Sperbeck
After TCP 3WHS became lockless, we should not attempt cgroup games
from sk_clone_lock() since listener/cgroup might be already gone.
Move this business to inet_csk_accept() where we have
the guarantee both parent and child exist.
Many thanks to John Sperbeck for spotting these issues
Eric Dumazet (2):
net: memcontrol: defer call to mem_cgroup_sk_alloc()
net: defer call to cgroup_sk_alloc()
kernel/cgroup/cgroup.c | 11 -----------
mm/memcontrol.c | 15 ---------------
net/core/sock.c | 8 +++++---
net/ipv4/inet_connection_sock.c | 6 ++++++
4 files changed, 11 insertions(+), 29 deletions(-)
--
2.14.2.920.gcf0c67979c-goog
^ permalink raw reply
* Re: [PATCH v2 net-next 06/12] qed: Add LL2 slowpath handling
From: David Miller @ 2017-10-09 4:40 UTC (permalink / raw)
To: Michal.Kalderon-YGCgFSpz5w/QT0dZR+AlfA
Cc: netdev-u79uwXL29TY76Z2rM5mHXA, linux-rdma-u79uwXL29TY76Z2rM5mHXA,
dledford-H+wXaHxf7aLQT0dZR+AlfA,
Ariel.Elior-YGCgFSpz5w/QT0dZR+AlfA
In-Reply-To: <CY1PR0701MB20128130D21FD3C54E45B5A188720-UpKza+2NMNLHMJvQ0dyT705OhdzP3rhOnBOFsp37pqbUKgpGm//BTAC/G2K4zDHf@public.gmane.org>
From: "Kalderon, Michal" <Michal.Kalderon-YGCgFSpz5w/QT0dZR+AlfA@public.gmane.org>
Date: Tue, 3 Oct 2017 18:05:32 +0000
> From: David Miller <davem-fT/PcQaiUtIeIZ0/mPfg9Q@public.gmane.org>
> Sent: Tuesday, October 3, 2017 8:17 PM
>>> @@ -423,6 +423,41 @@ static void qed_ll2_rxq_parse_reg(struct qed_hwfn *p_hwfn,
>>> }
>>>
>>> static int
>>> +qed_ll2_handle_slowpath(struct qed_hwfn *p_hwfn,
>>> + struct qed_ll2_info *p_ll2_conn,
>>> + union core_rx_cqe_union *p_cqe,
>>> + unsigned long *p_lock_flags)
>>> +{
>>...
>>> + spin_unlock_irqrestore(&p_rx->lock, *p_lock_flags);
>>> +
>>
>>You can't drop this lock.
>>
>>Another thread can enter the loop of our caller and process RX queue
>>entries, then we would return from here and try to process the same
>>entries again.
>
> The lock is there to synchronize access to chains between qed_ll2_rxq_completion
> and qed_ll2_post_rx_buffer. qed_ll2_rxq_completion can't be called from
> different threads, the light l2 uses the single sp status block we have.
> The reason we release the lock is to avoid a deadlock where as a result of calling
> upper-layer driver it will potentially post additional rx-buffers.
Ok, please repost this patch series.
Thanks.
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply
* Re: [PATCH] qed: Delete redundant check on dcb_app priority
From: David Miller @ 2017-10-09 4:21 UTC (permalink / raw)
To: chris.gekas; +Cc: Ariel.Elior, everest-linux-l2, netdev, linux-kernel
In-Reply-To: <1507502807-19767-1-git-send-email-chris.gekas@gmail.com>
From: Christos Gkekas <chris.gekas@gmail.com>
Date: Sun, 8 Oct 2017 23:46:47 +0100
> dcb_app priority is unsigned thus checking whether it is less than zero
> is redundant.
>
> Signed-off-by: Christos Gkekas <chris.gekas@gmail.com>
Applied to net-next.
^ permalink raw reply
* Re: [PATCH v2] net/core: Fix BUG to BUG_ON conditionals.
From: David Miller @ 2017-10-09 4:20 UTC (permalink / raw)
To: devtimhansen
Cc: willemb, edumazet, soheil, elena.reshetova, pabeni, tom, Jason,
fw, netdev, linux-kernel, alexander.levin
In-Reply-To: <20171008200338.2noygmfbdri6lc4y@debian>
From: Tim Hansen <devtimhansen@gmail.com>
Date: Sun, 8 Oct 2017 16:03:38 -0400
> Mistakenly sent the patch previously with a missing semicolon.
> Apologies.
>
> Fix BUG() calls to use BUG_ON(conditional) macros.
>
> This was found using make coccicheck M=net/core on linux next
> tag next-20170929
>
> Signed-off-by: Tim Hansen <devtimhansen@gmail.com>
You're going to have to submit this new version again, the way you
replied to the original patch caused the fixed version to not get
queued up in patchwork properly.
^ permalink raw reply
* Re: [PATCH] net: ethernet: stmmac: Clean up dead code
From: David Miller @ 2017-10-09 4:19 UTC (permalink / raw)
To: chris.gekas; +Cc: peppe.cavallaro, alexandre.torgue, netdev, linux-kernel
In-Reply-To: <1507490029-26406-1-git-send-email-chris.gekas@gmail.com>
From: Christos Gkekas <chris.gekas@gmail.com>
Date: Sun, 8 Oct 2017 20:13:49 +0100
> Many macros in dwmac-ipq806x are unused and should be removed.
> Moreover gmac->id is an unsigned variable and therefore checking
> whether it is less than zero is redundant.
>
> Signed-off-by: Christos Gkekas <chris.gekas@gmail.com>
Applied.
^ permalink raw reply
* Re: [PATCH net-next 0/6] ipv6: ipv6_dev_get_saddr() rcu works
From: David Miller @ 2017-10-09 4:17 UTC (permalink / raw)
To: edumazet; +Cc: netdev, eric.dumazet, yoshfuji
In-Reply-To: <20171008023028.32071-1-edumazet@google.com>
From: Eric Dumazet <edumazet@google.com>
Date: Sat, 7 Oct 2017 19:30:22 -0700
> Sending IPv6 udp packets on non connected sockets is quite slow,
> because ipv6_dev_get_saddr() is still using an rwlock and silly
> references games on ifa.
>
> Tested:
>
> $ ./super_netperf 16 -H 4444::555:0786 -l 2000 -t UDP_STREAM -- -m 100 &
> [1] 12527
>
> Performance is boosted from 2.02 Mpps to 4.28 Mpps
Awesome, series applied, thanks!
^ permalink raw reply
* Re: [PATCH] net: make ->ndo_get_phys_port_name accept 32-bit len
From: David Miller @ 2017-10-09 4:14 UTC (permalink / raw)
To: jakub.kicinski
Cc: adobriyan, netdev, michael.chan, saeedm, simon.horman, jiri,
ecree, vivien.didelot
In-Reply-To: <20171008182545.171498a2@cakuba.netronome.com>
From: Jakub Kicinski <jakub.kicinski@netronome.com>
Date: Sun, 8 Oct 2017 18:25:45 -0700
> On Sun, 8 Oct 2017 01:19:17 +0300, Alexey Dobriyan wrote:
>> Buffer length passed into this hook is always IFNAMSIZ which is 16.
>>
>> Code savings on x86_64:
>>
>> add/remove: 0/0 grow/shrink: 1/9 up/down: 2/-45 (-43)
>> function old new delta
>> rocker_cmd_get_port_settings_phys_name_proc 179 181 +2
>> rocker_port_get_phys_port_name 62 61 -1
>> mlxsw_sx_port_get_phys_port_name 54 50 -4
>> mlx5e_rep_get_phys_port_name 61 57 -4
>> efx_get_phys_port_name 50 46 -4
>> dsa_slave_get_phys_port_name 54 50 -4
>> bnxt_vf_rep_get_phys_port_name 69 65 -4
>> bnxt_get_phys_port_name 70 65 -5
>> mlxsw_sp_port_get_phys_port_name 116 107 -9
>> nfp_port_get_phys_port_name 180 170 -10
>>
>> Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
>
> I don't think the gains justify the additional burden on backports.
Yeah I agree, this one is not really worth the pain.
^ permalink raw reply
* Re: [net 1/1] tipc: Unclone message at secondary destination lookup
From: David Miller @ 2017-10-09 4:13 UTC (permalink / raw)
To: jon.maloy; +Cc: ying.xue, netdev, tipc-discussion
In-Reply-To: <1507381640-29340-1-git-send-email-jon.maloy@ericsson.com>
From: Jon Maloy <jon.maloy@ericsson.com>
Date: Sat, 7 Oct 2017 15:07:20 +0200
> When a bundling message is received, the function tipc_link_input()
> calls function tipc_msg_extract() to unbundle all inner messages of
> the bundling message before adding them to input queue.
>
> The function tipc_msg_extract() just clones all inner skb for all
> inner messagges from the bundling skb. This means that the skb
> headroom of an inner message overlaps with the data part of the
> preceding message in the bundle.
>
> If the message in question is a name addressed message, it may be
> subject to a secondary destination lookup, and eventually be sent out
> on one of the interfaces again. But, since what is perceived as headroom
> by the device driver in reality is the last bytes of the preceding
> message in the bundle, the latter will be overwritten by the MAC
> addresses of the L2 header. If the preceding message has not yet been
> consumed by the user, it will evenually be delivered with corrupted
> contents.
>
> This commit fixes this by uncloning all messages passing through the
> function tipc_msg_lookup_dest(), hence ensuring that the headroom
> is always valid when the message is passed on.
>
> Signed-off-by: Tung Nguyen <tung.q.nguyen@dektech.com.au>
> Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Applied.
------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
^ permalink raw reply
* Re: [net 1/1] tipc: correct initialization of skb list
From: David Miller @ 2017-10-09 4:13 UTC (permalink / raw)
To: jon.maloy; +Cc: netdev, parthasarathy.bhuvaragan, ying.xue, tipc-discussion
In-Reply-To: <1507379569-26507-1-git-send-email-jon.maloy@ericsson.com>
From: Jon Maloy <jon.maloy@ericsson.com>
Date: Sat, 7 Oct 2017 14:32:49 +0200
> We change the initialization of the skb transmit buffer queues
> in the functions tipc_bcast_xmit() and tipc_rcast_xmit() to also
> initialize their spinlocks. This is needed because we may, during
> error conditions, need to call skb_queue_purge() on those queues
> further down the stack.
>
> Signed-off-by: Jon Maloy <jon.maloy@ericsson.com>
Applied.
^ permalink raw reply
* Re: [PATCH net-next v6 0/3] bridge: neigh msg proxy and flood suppression support
From: David Miller @ 2017-10-09 4:12 UTC (permalink / raw)
To: roopa; +Cc: netdev, nikolay, stephen, makita.toshiaki, bridge
In-Reply-To: <1507353159-46814-1-git-send-email-roopa@cumulusnetworks.com>
From: Roopa Prabhu <roopa@cumulusnetworks.com>
Date: Fri, 6 Oct 2017 22:12:36 -0700
> From: Roopa Prabhu <roopa@cumulusnetworks.com>
>
> This series implements arp and nd suppression in the bridge
> driver for ethernet vpns. It implements rfc7432, section 10
> https://tools.ietf.org/html/rfc7432#section-10
> for ethernet VPN deployments. It is similar to the existing
> BR_PROXYARP* flags but has a few semantic differences to conform
> to EVPN standard. Unlike the existing flags, this new flag suppresses
> flood of all neigh discovery packets (arp and nd) to tunnel ports.
> Supports both vlan filtering and non-vlan filtering bridges.
>
> In case of EVPN, it is mainly used to avoid flooding
> of arp and nd packets to tunnel ports like vxlan.
>
> v2 : rebase to latest + address some optimization feedback from Nikolay.
> v3 : fix kbuild reported build errors with CONFIG_INET off
> v4 : simplify port flag mask as suggested by stephen
> v5 : address some feedback from Toshiaki
> v6 : some v5 cleanups in nd suppress (keep it consistent with arp suppress)
Series applied, thank you.
^ permalink raw reply
* Re: [PATCH net-next] ipv6: fix a BUG in rt6_get_pcpu_route()
From: David Miller @ 2017-10-09 4:09 UTC (permalink / raw)
To: eric.dumazet; +Cc: idosch, netdev, weiwan, mlxsw
In-Reply-To: <1507522038.31614.3.camel@edumazet-glaptop3.roam.corp.google.com>
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Sun, 08 Oct 2017 21:07:18 -0700
> From: Eric Dumazet <edumazet@google.com>
>
> Ido reported following splat and provided a patch.
>
> [ 122.221814] BUG: using smp_processor_id() in preemptible [00000000] code: sshd/2672
> [ 122.221845] caller is debug_smp_processor_id+0x17/0x20
> [ 122.221866] CPU: 0 PID: 2672 Comm: sshd Not tainted 4.14.0-rc3-idosch-next-custom #639
> [ 122.221880] Hardware name: Mellanox Technologies Ltd. MSN2100-CB2FO/SA001017, BIOS 5.6.5 06/07/2016
> [ 122.221893] Call Trace:
> [ 122.221919] dump_stack+0xb1/0x10c
> [ 122.221946] ? _atomic_dec_and_lock+0x124/0x124
> [ 122.221974] ? ___ratelimit+0xfe/0x240
> [ 122.222020] check_preemption_disabled+0x173/0x1b0
> [ 122.222060] debug_smp_processor_id+0x17/0x20
> [ 122.222083] ip6_pol_route+0x1482/0x24a0
> ...
>
> I believe we can simplify this code path a bit, since we no longer
> hold a read_lock and need to release it to avoid a dead lock.
>
> By disabling BH, we make sure we'll prevent code re-entry and
> rt6_get_pcpu_route()/rt6_make_pcpu_route() run on the same cpu.
>
> Fixes: 66f5d6ce53e6 ("ipv6: replace rwlock with rcu and spinlock in fib6_table")
> Reported-by: Ido Schimmel <idosch@mellanox.com>
> Signed-off-by: Eric Dumazet <edumazet@google.com>
> Tested-by: Ido Schimmel <idosch@mellanox.com>
Applied, thanks Eric.
^ permalink raw reply
* Re: [pull request][for-next 0/9] Mellanox, mlx5 updates 2017-10-06
From: David Miller @ 2017-10-09 4:08 UTC (permalink / raw)
To: saeedm-VPRAkNaXOzVWk0Htik3J/w
Cc: dledford-H+wXaHxf7aLQT0dZR+AlfA, netdev-u79uwXL29TY76Z2rM5mHXA,
linux-rdma-u79uwXL29TY76Z2rM5mHXA, leonro-VPRAkNaXOzVWk0Htik3J/w
In-Reply-To: <20171006233749.25545-1-saeedm-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
From: Saeed Mahameed <saeedm-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
Date: Fri, 6 Oct 2017 16:37:40 -0700
> This series includes some shared code updates for kernel 4.15 to both
> net-next and rdma-next trees.
I've pulled this into net-next, thanks Saeed.
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply
* [PATCH net-next] ipv6: fix a BUG in rt6_get_pcpu_route()
From: Eric Dumazet @ 2017-10-09 4:07 UTC (permalink / raw)
To: Ido Schimmel; +Cc: netdev, davem, weiwan, mlxsw
In-Reply-To: <1507478633.14419.30.camel@edumazet-glaptop3.roam.corp.google.com>
From: Eric Dumazet <edumazet@google.com>
Ido reported following splat and provided a patch.
[ 122.221814] BUG: using smp_processor_id() in preemptible [00000000] code: sshd/2672
[ 122.221845] caller is debug_smp_processor_id+0x17/0x20
[ 122.221866] CPU: 0 PID: 2672 Comm: sshd Not tainted 4.14.0-rc3-idosch-next-custom #639
[ 122.221880] Hardware name: Mellanox Technologies Ltd. MSN2100-CB2FO/SA001017, BIOS 5.6.5 06/07/2016
[ 122.221893] Call Trace:
[ 122.221919] dump_stack+0xb1/0x10c
[ 122.221946] ? _atomic_dec_and_lock+0x124/0x124
[ 122.221974] ? ___ratelimit+0xfe/0x240
[ 122.222020] check_preemption_disabled+0x173/0x1b0
[ 122.222060] debug_smp_processor_id+0x17/0x20
[ 122.222083] ip6_pol_route+0x1482/0x24a0
...
I believe we can simplify this code path a bit, since we no longer
hold a read_lock and need to release it to avoid a dead lock.
By disabling BH, we make sure we'll prevent code re-entry and
rt6_get_pcpu_route()/rt6_make_pcpu_route() run on the same cpu.
Fixes: 66f5d6ce53e6 ("ipv6: replace rwlock with rcu and spinlock in fib6_table")
Reported-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Tested-by: Ido Schimmel <idosch@mellanox.com>
---
net/ipv6/route.c | 26 ++++++--------------------
1 file changed, 6 insertions(+), 20 deletions(-)
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 399d1bceec4a6e6736c367e706dd2acbd4093d58..606e80325b21c0e10a02e9c7d5b3fcfbfc26a003 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -1136,15 +1136,7 @@ static struct rt6_info *rt6_make_pcpu_route(struct rt6_info *rt)
dst_hold(&pcpu_rt->dst);
p = this_cpu_ptr(rt->rt6i_pcpu);
prev = cmpxchg(p, NULL, pcpu_rt);
- if (prev) {
- /* If someone did it before us, return prev instead */
- /* release refcnt taken by ip6_rt_pcpu_alloc() */
- dst_release_immediate(&pcpu_rt->dst);
- /* release refcnt taken by above dst_hold() */
- dst_release_immediate(&pcpu_rt->dst);
- dst_hold(&prev->dst);
- pcpu_rt = prev;
- }
+ BUG_ON(prev);
rt6_dst_from_metrics_check(pcpu_rt);
return pcpu_rt;
@@ -1739,31 +1731,25 @@ struct rt6_info *ip6_pol_route(struct net *net, struct fib6_table *table,
struct rt6_info *pcpu_rt;
dst_use_noref(&rt->dst, jiffies);
+ local_bh_disable();
pcpu_rt = rt6_get_pcpu_route(rt);
- if (pcpu_rt) {
- rcu_read_unlock();
- } else {
+ if (!pcpu_rt) {
/* atomic_inc_not_zero() is needed when using rcu */
if (atomic_inc_not_zero(&rt->rt6i_ref)) {
- /* We have to do the read_unlock first
- * because rt6_make_pcpu_route() may trigger
- * ip6_dst_gc() which will take the write_lock.
- *
- * No dst_hold() on rt is needed because grabbing
+ /* No dst_hold() on rt is needed because grabbing
* rt->rt6i_ref makes sure rt can't be released.
*/
- rcu_read_unlock();
pcpu_rt = rt6_make_pcpu_route(rt);
rt6_release(rt);
} else {
/* rt is already removed from tree */
- rcu_read_unlock();
pcpu_rt = net->ipv6.ip6_null_entry;
dst_hold(&pcpu_rt->dst);
}
}
-
+ local_bh_enable();
+ rcu_read_unlock();
trace_fib6_table_lookup(net, pcpu_rt, table->tb6_id, fl6);
return pcpu_rt;
}
^ permalink raw reply related
* [PATCH net-next 15/15] nfp: bpf: pass dst register to ld_field instruction
From: Jakub Kicinski @ 2017-10-09 4:04 UTC (permalink / raw)
To: netdev; +Cc: oss-drivers, Jakub Kicinski
In-Reply-To: <20171009040417.22172-1-jakub.kicinski@netronome.com>
ld_field instruction is a bit special because the encoding uses
two source registers and one of them becomes the output. We do
need to pass the dst register to our encoding helpers though,
otherwise the "write both banks" flag will not be observed.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
---
drivers/net/ethernet/netronome/nfp/bpf/jit.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/netronome/nfp/bpf/jit.c b/drivers/net/ethernet/netronome/nfp/bpf/jit.c
index f68052367db7..13148f30fc4c 100644
--- a/drivers/net/ethernet/netronome/nfp/bpf/jit.c
+++ b/drivers/net/ethernet/netronome/nfp/bpf/jit.c
@@ -408,7 +408,8 @@ emit_ld_field_any(struct nfp_prog *nfp_prog, enum shf_sc sc, u8 shift,
struct nfp_insn_re_regs reg;
int err;
- err = swreg_to_restricted(reg_none(), dst, src, ®, true);
+ /* Note: ld_field is special as it uses one of the src regs as dst */
+ err = swreg_to_restricted(dst, dst, src, ®, true);
if (err) {
nfp_prog->error = err;
return;
--
2.14.1
^ permalink raw reply related
* [PATCH net-next 14/15] nfp: bpf: byte swap the instructions
From: Jakub Kicinski @ 2017-10-09 4:04 UTC (permalink / raw)
To: netdev; +Cc: oss-drivers, Jakub Kicinski
In-Reply-To: <20171009040417.22172-1-jakub.kicinski@netronome.com>
Device expects the instructions in little endian. Make sure we
byte swap on big endian hosts.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
---
drivers/net/ethernet/netronome/nfp/bpf/jit.c | 6 ++++--
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ethernet/netronome/nfp/bpf/jit.c b/drivers/net/ethernet/netronome/nfp/bpf/jit.c
index e0600d037773..f68052367db7 100644
--- a/drivers/net/ethernet/netronome/nfp/bpf/jit.c
+++ b/drivers/net/ethernet/netronome/nfp/bpf/jit.c
@@ -1725,7 +1725,7 @@ static int nfp_bpf_optimize(struct nfp_prog *nfp_prog)
return 0;
}
-static int nfp_bpf_ustore_calc(struct nfp_prog *nfp_prog)
+static int nfp_bpf_ustore_calc(struct nfp_prog *nfp_prog, __le64 *ustore)
{
int i;
@@ -1737,6 +1737,8 @@ static int nfp_bpf_ustore_calc(struct nfp_prog *nfp_prog)
return err;
nfp_prog->prog[i] = nfp_ustore_calc_ecc_insn(nfp_prog->prog[i]);
+
+ ustore[i] = cpu_to_le64(nfp_prog->prog[i]);
}
return 0;
@@ -1796,7 +1798,7 @@ nfp_bpf_jit(struct bpf_prog *filter, void *prog_mem,
goto out;
}
- ret = nfp_bpf_ustore_calc(nfp_prog);
+ ret = nfp_bpf_ustore_calc(nfp_prog, (__force __le64 *)prog_mem);
res->n_instr = nfp_prog->prog_len;
res->dense_mode = false;
--
2.14.1
^ permalink raw reply related
* [PATCH net-next 12/15] nfp: bpf: calculate code store ECC
From: Jakub Kicinski @ 2017-10-09 4:04 UTC (permalink / raw)
To: netdev; +Cc: oss-drivers, Jakub Kicinski
In-Reply-To: <20171009040417.22172-1-jakub.kicinski@netronome.com>
In the initial PoC firmware I simply disabled ECC on the instruction
store. Do the ECC calculation for generated instructions in the driver.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
---
drivers/net/ethernet/netronome/nfp/bpf/jit.c | 20 +++++++++++++++
drivers/net/ethernet/netronome/nfp/nfp_asm.c | 37 ++++++++++++++++++++++++++++
drivers/net/ethernet/netronome/nfp/nfp_asm.h | 3 +++
3 files changed, 60 insertions(+)
diff --git a/drivers/net/ethernet/netronome/nfp/bpf/jit.c b/drivers/net/ethernet/netronome/nfp/bpf/jit.c
index 9b6c98ccebfe..f4aedc89bfc8 100644
--- a/drivers/net/ethernet/netronome/nfp/bpf/jit.c
+++ b/drivers/net/ethernet/netronome/nfp/bpf/jit.c
@@ -1715,6 +1715,23 @@ static int nfp_bpf_optimize(struct nfp_prog *nfp_prog)
return 0;
}
+static int nfp_bpf_ustore_calc(struct nfp_prog *nfp_prog)
+{
+ int i;
+
+ for (i = 0; i < nfp_prog->prog_len; i++) {
+ int err;
+
+ err = nfp_ustore_check_valid_no_ecc(nfp_prog->prog[i]);
+ if (err)
+ return err;
+
+ nfp_prog->prog[i] = nfp_ustore_calc_ecc_insn(nfp_prog->prog[i]);
+ }
+
+ return 0;
+}
+
/**
* nfp_bpf_jit() - translate BPF code into NFP assembly
* @filter: kernel BPF filter struct
@@ -1766,8 +1783,11 @@ nfp_bpf_jit(struct bpf_prog *filter, void *prog_mem,
pr_err("Translation failed with error %d (translated: %u)\n",
ret, nfp_prog->n_translated);
ret = -EINVAL;
+ goto out;
}
+ ret = nfp_bpf_ustore_calc(nfp_prog);
+
res->n_instr = nfp_prog->prog_len;
res->dense_mode = false;
out:
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_asm.c b/drivers/net/ethernet/netronome/nfp/nfp_asm.c
index 1decc638ea6f..de76e7444fc2 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_asm.c
+++ b/drivers/net/ethernet/netronome/nfp/nfp_asm.c
@@ -215,3 +215,40 @@ int swreg_to_restricted(swreg dst, swreg lreg, swreg rreg,
return 0;
}
+
+#define NFP_USTORE_ECC_POLY_WORDS 7
+#define NFP_USTORE_OP_BITS 45
+
+static const u64 nfp_ustore_ecc_polynomials[NFP_USTORE_ECC_POLY_WORDS] = {
+ 0x0ff800007fffULL,
+ 0x11f801ff801fULL,
+ 0x1e387e0781e1ULL,
+ 0x17cb8e388e22ULL,
+ 0x1af5b2c93244ULL,
+ 0x1f56d5525488ULL,
+ 0x0daf69a46910ULL,
+};
+
+static bool parity(u64 value)
+{
+ return hweight64(value) & 1;
+}
+
+int nfp_ustore_check_valid_no_ecc(u64 insn)
+{
+ if (insn & ~GENMASK_ULL(NFP_USTORE_OP_BITS, 0))
+ return -EINVAL;
+
+ return 0;
+}
+
+u64 nfp_ustore_calc_ecc_insn(u64 insn)
+{
+ u8 ecc = 0;
+ int i;
+
+ for (i = 0; i < NFP_USTORE_ECC_POLY_WORDS; i++)
+ ecc |= parity(nfp_ustore_ecc_polynomials[i] & insn) << i;
+
+ return insn | (u64)ecc << NFP_USTORE_OP_BITS;
+}
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_asm.h b/drivers/net/ethernet/netronome/nfp/nfp_asm.h
index 40a51a45afd7..d95087e5fb73 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_asm.h
+++ b/drivers/net/ethernet/netronome/nfp/nfp_asm.h
@@ -362,4 +362,7 @@ int swreg_to_unrestricted(swreg dst, swreg lreg, swreg rreg,
int swreg_to_restricted(swreg dst, swreg lreg, swreg rreg,
struct nfp_insn_re_regs *reg, bool has_imm8);
+int nfp_ustore_check_valid_no_ecc(u64 insn);
+u64 nfp_ustore_calc_ecc_insn(u64 insn);
+
#endif
--
2.14.1
^ permalink raw reply related
* [PATCH net-next 13/15] nfp: bpf: pad code with valid nops
From: Jakub Kicinski @ 2017-10-09 4:04 UTC (permalink / raw)
To: netdev; +Cc: oss-drivers, Jakub Kicinski
In-Reply-To: <20171009040417.22172-1-jakub.kicinski@netronome.com>
We need to append up to 8 nops after last instruction to make
sure the CPU will not fetch garbage instructions with invalid
ECC if the code store was not initialized.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
---
drivers/net/ethernet/netronome/nfp/bpf/jit.c | 12 +++++++++++-
drivers/net/ethernet/netronome/nfp/nfp_asm.h | 2 ++
2 files changed, 13 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/netronome/nfp/bpf/jit.c b/drivers/net/ethernet/netronome/nfp/bpf/jit.c
index f4aedc89bfc8..e0600d037773 100644
--- a/drivers/net/ethernet/netronome/nfp/bpf/jit.c
+++ b/drivers/net/ethernet/netronome/nfp/bpf/jit.c
@@ -426,6 +426,11 @@ emit_ld_field(struct nfp_prog *nfp_prog, swreg dst, u8 bmask, swreg src,
emit_ld_field_any(nfp_prog, sc, shift, dst, bmask, src, false);
}
+static void emit_nop(struct nfp_prog *nfp_prog)
+{
+ __emit_immed(nfp_prog, UR_REG_IMM, UR_REG_IMM, 0, 0, 0, 0, 0, 0, 0);
+}
+
/* --- Wrappers --- */
static bool pack_immed(u32 imm, u16 *val, enum immed_shift *shift)
{
@@ -1550,7 +1555,7 @@ static void nfp_outro(struct nfp_prog *nfp_prog)
static int nfp_translate(struct nfp_prog *nfp_prog)
{
struct nfp_insn_meta *meta;
- int err;
+ int i, err;
nfp_intro(nfp_prog);
if (nfp_prog->error)
@@ -1582,6 +1587,11 @@ static int nfp_translate(struct nfp_prog *nfp_prog)
if (nfp_prog->error)
return nfp_prog->error;
+ for (i = 0; i < NFP_USTORE_PREFETCH_WINDOW; i++)
+ emit_nop(nfp_prog);
+ if (nfp_prog->error)
+ return nfp_prog->error;
+
return nfp_fixup_branches(nfp_prog);
}
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_asm.h b/drivers/net/ethernet/netronome/nfp/nfp_asm.h
index d95087e5fb73..c4c18dd5630a 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_asm.h
+++ b/drivers/net/ethernet/netronome/nfp/nfp_asm.h
@@ -362,6 +362,8 @@ int swreg_to_unrestricted(swreg dst, swreg lreg, swreg rreg,
int swreg_to_restricted(swreg dst, swreg lreg, swreg rreg,
struct nfp_insn_re_regs *reg, bool has_imm8);
+#define NFP_USTORE_PREFETCH_WINDOW 8
+
int nfp_ustore_check_valid_no_ecc(u64 insn);
u64 nfp_ustore_calc_ecc_insn(u64 insn);
--
2.14.1
^ permalink raw reply related
* [PATCH net-next 11/15] nfp: bpf: move to datapath ABI version 2
From: Jakub Kicinski @ 2017-10-09 4:04 UTC (permalink / raw)
To: netdev; +Cc: oss-drivers, Jakub Kicinski
In-Reply-To: <20171009040417.22172-1-jakub.kicinski@netronome.com>
Datapath ABI version 2 stores the packet information in LMEM
instead of NNRs. We also have strict restrictions on which
GPRs we can use. Only GPRs 0-23 are reserved for BPF.
Adjust the static register locations and "ABI" registers.
Note that packet length is packed with other info so we have
to extract it into one of the scratch registers, OTOH since
LMEM can be used in restricted operands we don't have to
extract packet pointer.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
---
drivers/net/ethernet/netronome/nfp/bpf/jit.c | 19 +++++++++---------
drivers/net/ethernet/netronome/nfp/bpf/main.h | 24 ++++++++++++++---------
drivers/net/ethernet/netronome/nfp/nfp_net_ctrl.h | 2 +-
3 files changed, 26 insertions(+), 19 deletions(-)
diff --git a/drivers/net/ethernet/netronome/nfp/bpf/jit.c b/drivers/net/ethernet/netronome/nfp/bpf/jit.c
index d7dc19feba8d..9b6c98ccebfe 100644
--- a/drivers/net/ethernet/netronome/nfp/bpf/jit.c
+++ b/drivers/net/ethernet/netronome/nfp/bpf/jit.c
@@ -526,22 +526,22 @@ construct_data_ind_ld(struct nfp_prog *nfp_prog, u16 offset,
emit_alu(nfp_prog, imm_a(nfp_prog),
imm_a(nfp_prog), ALU_OP_ADD, reg_imm(size));
emit_alu(nfp_prog, reg_none(),
- NFP_BPF_ABI_LEN, ALU_OP_SUB, imm_a(nfp_prog));
+ plen_reg(nfp_prog), ALU_OP_SUB, imm_a(nfp_prog));
wrp_br_special(nfp_prog, BR_BLO, OP_BR_GO_ABORT);
/* Load data */
emit_cmd(nfp_prog, CMD_TGT_READ8, CMD_MODE_32b, 0,
- pkt_reg(nfp_prog), imm_b(nfp_prog), sz - 1, true);
+ pptr_reg(nfp_prog), imm_b(nfp_prog), sz - 1, true);
} else {
/* Check packet length */
tmp_reg = ur_load_imm_any(nfp_prog, offset + size,
imm_a(nfp_prog));
emit_alu(nfp_prog, reg_none(),
- NFP_BPF_ABI_LEN, ALU_OP_SUB, tmp_reg);
+ plen_reg(nfp_prog), ALU_OP_SUB, tmp_reg);
wrp_br_special(nfp_prog, BR_BLO, OP_BR_GO_ABORT);
/* Load data */
tmp_reg = re_load_imm_any(nfp_prog, offset, imm_b(nfp_prog));
emit_cmd(nfp_prog, CMD_TGT_READ8, CMD_MODE_32b, 0,
- pkt_reg(nfp_prog), tmp_reg, sz - 1, true);
+ pptr_reg(nfp_prog), tmp_reg, sz - 1, true);
}
i = 0;
@@ -1024,7 +1024,7 @@ static int mem_ldx4_skb(struct nfp_prog *nfp_prog, struct nfp_insn_meta *meta)
{
if (meta->insn.off == offsetof(struct sk_buff, len))
emit_alu(nfp_prog, reg_both(meta->insn.dst_reg * 2),
- reg_none(), ALU_OP_NONE, NFP_BPF_ABI_LEN);
+ reg_none(), ALU_OP_NONE, plen_reg(nfp_prog));
else
return -EOPNOTSUPP;
@@ -1039,12 +1039,12 @@ static int mem_ldx4_xdp(struct nfp_prog *nfp_prog, struct nfp_insn_meta *meta)
meta->insn.off != offsetof(struct xdp_md, data_end))
return -EOPNOTSUPP;
- emit_alu(nfp_prog, dst, reg_none(), ALU_OP_NONE, NFP_BPF_ABI_PKT);
+ emit_alu(nfp_prog, dst, reg_none(), ALU_OP_NONE, pptr_reg(nfp_prog));
if (meta->insn.off == offsetof(struct xdp_md, data))
return 0;
- emit_alu(nfp_prog, dst, dst, ALU_OP_ADD, NFP_BPF_ABI_LEN);
+ emit_alu(nfp_prog, dst, dst, ALU_OP_ADD, plen_reg(nfp_prog));
return 0;
}
@@ -1403,8 +1403,9 @@ static int nfp_fixup_branches(struct nfp_prog *nfp_prog)
static void nfp_intro(struct nfp_prog *nfp_prog)
{
- emit_alu(nfp_prog, pkt_reg(nfp_prog),
- reg_none(), ALU_OP_NONE, NFP_BPF_ABI_PKT);
+ wrp_immed(nfp_prog, plen_reg(nfp_prog), GENMASK(13, 0));
+ emit_alu(nfp_prog, plen_reg(nfp_prog),
+ plen_reg(nfp_prog), ALU_OP_AND, pv_len(nfp_prog));
}
static void nfp_outro_tc_legacy(struct nfp_prog *nfp_prog)
diff --git a/drivers/net/ethernet/netronome/nfp/bpf/main.h b/drivers/net/ethernet/netronome/nfp/bpf/main.h
index 7d959757a51a..b7a112acbdb7 100644
--- a/drivers/net/ethernet/netronome/nfp/bpf/main.h
+++ b/drivers/net/ethernet/netronome/nfp/bpf/main.h
@@ -54,9 +54,13 @@ enum br_special {
};
enum static_regs {
- STATIC_REG_PKT = 1,
-#define REG_PKT_BANK ALU_DST_A
- STATIC_REG_IMM = 2, /* Bank AB */
+ STATIC_REG_IMM = 21, /* Bank AB */
+ STATIC_REG_PKT_LEN = 22, /* Bank B */
+};
+
+enum pkt_vec {
+ PKT_VEC_PKT_LEN = 0,
+ PKT_VEC_PKT_PTR = 2,
};
enum nfp_bpf_action_type {
@@ -66,15 +70,17 @@ enum nfp_bpf_action_type {
NN_ACT_XDP,
};
-#define pkt_reg(np) reg_a((np)->regs_per_thread - STATIC_REG_PKT)
-#define imm_a(np) reg_a((np)->regs_per_thread - STATIC_REG_IMM)
-#define imm_b(np) reg_b((np)->regs_per_thread - STATIC_REG_IMM)
-#define imm_both(np) reg_both((np)->regs_per_thread - STATIC_REG_IMM)
+#define pv_len(np) reg_lm(1, PKT_VEC_PKT_LEN)
+#define pv_ctm_ptr(np) reg_lm(1, PKT_VEC_PKT_PTR)
+
+#define plen_reg(np) reg_b(STATIC_REG_PKT_LEN)
+#define pptr_reg(np) pv_ctm_ptr(np)
+#define imm_a(np) reg_a(STATIC_REG_IMM)
+#define imm_b(np) reg_b(STATIC_REG_IMM)
+#define imm_both(np) reg_both(STATIC_REG_IMM)
#define NFP_BPF_ABI_FLAGS reg_imm(0)
#define NFP_BPF_ABI_FLAG_MARK 1
-#define NFP_BPF_ABI_PKT reg_nnr(2)
-#define NFP_BPF_ABI_LEN reg_nnr(3)
struct nfp_prog;
struct nfp_insn_meta;
diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_ctrl.h b/drivers/net/ethernet/netronome/nfp/nfp_net_ctrl.h
index b0a452ba9039..782d452e0fc2 100644
--- a/drivers/net/ethernet/netronome/nfp/nfp_net_ctrl.h
+++ b/drivers/net/ethernet/netronome/nfp/nfp_net_ctrl.h
@@ -255,7 +255,7 @@
* @NFP_NET_CFG_BPF_ADDR: DMA address of the buffer with JITed BPF code
*/
#define NFP_NET_CFG_BPF_ABI 0x0080
-#define NFP_NET_BPF_ABI 1
+#define NFP_NET_BPF_ABI 2
#define NFP_NET_CFG_BPF_CAP 0x0081
#define NFP_NET_BPF_CAP_RELO (1 << 0) /* seamless reload */
#define NFP_NET_CFG_BPF_MAX_LEN 0x0082
--
2.14.1
^ permalink raw reply related
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