* Re: [PATCH] SUNRPC: Move inline xprt_alloc_xid() up to fix compiler warning
From: Geert Uytterhoeven @ 2018-06-15 8:23 UTC (permalink / raw)
To: Chuck Lever
Cc: Bruce Fields, jlayton, trond.myklebust, Anna Schumaker,
David S. Miller, open list:NFS, SUNRPC, AND..., netdev,
Linux Kernel Mailing List
In-Reply-To: <675726D3-9273-4051-B38F-6377B8445A27@oracle.com>
Hi Chuck,
On Thu, Jun 14, 2018 at 7:20 PM Chuck Lever <chuck.lever@oracle.com> wrote:
> > On Jun 13, 2018, at 8:01 AM, Geert Uytterhoeven <geert@linux-m68k.org> wrote:
> >
> > With gcc 4.1.2:
> >
> > net/sunrpc/xprt.c:69: warning: ‘xprt_alloc_xid’ declared inline after being called
> > net/sunrpc/xprt.c:69: warning: previous declaration of ‘xprt_alloc_xid’ was here
> >
> > To fix this, move the function up, before its caller, and remove the no
> > longer needed forward declaration.
> >
> > Fixes: 37ac86c3a76c1136 ("SUNRPC: Initialize rpc_rqst outside of xprt->reserve_lock")
> > Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
> > ---
> > net/sunrpc/xprt.c | 11 +++++------
> > 1 file changed, 5 insertions(+), 6 deletions(-)
> >
> > diff --git a/net/sunrpc/xprt.c b/net/sunrpc/xprt.c
> > index 3c85af058227d14b..60a8b9f91cf94b54 100644
> > --- a/net/sunrpc/xprt.c
> > +++ b/net/sunrpc/xprt.c
> > @@ -66,7 +66,6 @@
> > * Local functions
> > */
> > static void xprt_init(struct rpc_xprt *xprt, struct net *net);
> > -static __be32 xprt_alloc_xid(struct rpc_xprt *xprt);
> > static void xprt_connect_status(struct rpc_task *task);
> > static int __xprt_get_cong(struct rpc_xprt *, struct rpc_task *);
> > static void __xprt_put_cong(struct rpc_xprt *, struct rpc_rqst *);
> > @@ -956,6 +955,11 @@ static void xprt_timer(struct rpc_task *task)
> > task->tk_status = 0;
> > }
> >
> > +static inline __be32 xprt_alloc_xid(struct rpc_xprt *xprt)
> > +{
> > + return (__force __be32)xprt->xid++;
> > +}
> > +
> > /**
> > * xprt_prepare_transmit - reserve the transport before sending a request
> > * @task: RPC task about to send a request
> > @@ -1296,11 +1300,6 @@ void xprt_retry_reserve(struct rpc_task *task)
> > xprt->ops->alloc_slot(xprt, task);
> > }
> >
> > -static inline __be32 xprt_alloc_xid(struct rpc_xprt *xprt)
> > -{
> > - return (__force __be32)xprt->xid++;
> > -}
> > -
>
> For code organization, we might want to keep xprt_alloc_xid
> together with xprt_init_xid. Would it be better to simply
> remove the "inline" directive from these two and let the
> compiler choose the best optimization?
That's an option, too.
> > static inline void xprt_init_xid(struct rpc_xprt *xprt)
> > {
> > xprt->xid = prandom_u32();
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
^ permalink raw reply
* Re: [RFC PATCH RESEND] tcp: avoid F-RTO if SACK and timestamps are disabled
From: Ilpo Järvinen @ 2018-06-15 8:05 UTC (permalink / raw)
To: Michal Kubecek; +Cc: Yuchung Cheng, netdev, Eric Dumazet, LKML
In-Reply-To: <20180614131801.hd474jgrhmtqzhag@unicorn.suse.cz>
[-- Attachment #1: Type: text/plain, Size: 2529 bytes --]
On Thu, 14 Jun 2018, Michal Kubecek wrote:
> On Thu, Jun 14, 2018 at 02:51:18PM +0300, Ilpo Järvinen wrote:
> > On Thu, 14 Jun 2018, Michal Kubecek wrote:
> > > On Thu, Jun 14, 2018 at 11:42:43AM +0300, Ilpo Järvinen wrote:
> > > > On Wed, 13 Jun 2018, Yuchung Cheng wrote:
> > > > > On Wed, Jun 13, 2018 at 9:55 AM, Michal Kubecek <mkubecek@suse.cz> wrote:
> > >
> > > AFAICS RFC 5682 is not explicit about this and offers multiple options.
> > > Anyway, this is not essential and in most of the customer provided
> > > captures, it wasn't the case.
> >
> > Lacking the new segments is essential for hiding the actual bug as the
> > trace would look weird otherwise with a burst of new data segments (due
> > to the other bug).
>
> The trace wouldn't look so nice but it can be reproduced even with more
> data to send. I've copied an example below. I couldn't find a really
> nice one quickly so that first few retransmits (17:22:13.865105 through
> 17:23:05.841105) are without new data but starting at 17:23:58.189150,
> you can see that sending new (previously unsent) data may not suffice to
> break the loop.
My point was that the new data segment bursts that occur if the sender
isn't application limited indicate that there's something going wrong
with FRTO. And that wrong is also what is causing that RTO loop because
the sender doesn't see the previous FRTO recovery on second RTO. With
my FRTO undo fix, (new_recovery || icsk->icsk_retransmits) will be false
and that will prevent the RTO loop.
> > > Normally, we would have timestamps (and even SACK). Without them, you
> > > cannot reliably recognize a dupack with changed window size from
> > > a spontaneous window update.
> >
> > No! The window should not update window on ACKs the receiver intends to
> > designate as "duplicate ACKs". That is not without some potential cost
> > though as it requires delaying window updates up to the next cumulative
> > ACK. In the non-SACK series one of the changes is fixing this for
> > non-SACK Linux TCP flows.
>
> That sounds like a reasonable change (at least at the first glance,
> I didn't think about it too deeply) but even if we fix Linux stack to
> behave like this, we cannot force everyone else to do the same.
Unfortunately I don't know what the other stacks besides Linux do. But
for Linux, the cause for the changing receiver window is the receiver
window auto-tuning and I'm not sure if other stacks have a similar
feature (or if that affects (almost) all ACKs like in Linux).
--
i.
^ permalink raw reply
* Re: [PATCH] optoe: driver to read/write SFP/QSFP EEPROMs
From: Andrew Lunn @ 2018-06-15 7:54 UTC (permalink / raw)
To: Don Bollinger
Cc: Tom Lendacky, Arnd Bergmann, Greg Kroah-Hartman, linux-kernel,
brandon_chuang, wally_wang, roy_lee, rick_burchett, quentin.chang,
steven.noble, jeffrey.townsend, scotte, roopa, David Ahern,
luke.williams, Guohan Lu, Russell King, netdev@vger.kernel.org
In-Reply-To: <20180615022652.t6oqpnwwvdmbooab@thebollingers.org>
> Actually this is better described by a third use case. The target
> switches are PHY-less (see various designs at
> www.compute.org/wiki/Networking/SpecsAndDesigns). The AS5712 for example
> says "The AS5712-54X is a PHY-Less design with the SFP+ and QSFP+
> connections directly attaching to the Serdes interfaces of the Broadcom
> BCM56854 720G Trident 2 switching silicon..."
We consider the SFP+ and QSFP+ as being the PHY. You need something to
control that PHY. Either it is firmware running in the switch, or it
is the Linux kernel, via PHYLINK.
> The i2c bus is muxed from the CPU to all of the {Q}SFP devices, which
> are set up as standard linux i2c devices
> (/sys/bus/i2c/devices/i2c-xxxx).
Having a standard i2c bus driver is correct. This is what PHYLINK
assumes. It knows about the different addresses the SFP uses on the
i2c bus.
> There is no MDIO bus between the CPU and the {Q}SFP devices.
There is no physical MDIO bus for SFP devices. If the SFP module
implements copper 1G, there is often MDIO tunnelled over i2c. PHYLINK
knows how to do this, and will instantiate a normal Linux MDIO bus
driver, and then you can use the Linux kernel copper PHY state
machines as normal.
> And, there isn't actually 'a wish to expose' the EEPROM data to linux
> (the kernel). It turns out that none of the NOS partners I'm working
> with use that data *in the kernel*. It is all managed from user space.
Ah. O.K. We can stop here then.
If you are using Linux as a boot loader, i doubt you will find any
network kernel developers who are willing to consider this driver. The
kernel community as decided switchdev is how the Linux kernel supports
switches. We are unlikely to add drivers for supporting user space
drivers of switches.
NACK.
Andrew
^ permalink raw reply
* Re: [PATCH 0/3] Use sbitmap instead of percpu_ida
From: Christoph Hellwig @ 2018-06-15 7:37 UTC (permalink / raw)
To: Matthew Wilcox
Cc: Juergen Gross, Jens Axboe, kvm, linux-scsi, netdev, linux-usb,
linux-kernel, virtualization, target-devel, qla2xxx-upstream,
linux1394-devel, Kent Overstreet
In-Reply-To: <20180612190545.10781-1-willy@infradead.org>
Btw, if you are on a spree to remove almost unused data structures
from target code, the lib/btree.c code is only used by the qla2xxx
target code, and doesn't really look like the best fit for it either.
^ permalink raw reply
* Re: BUG: KASAN: stack-out-of-bounds in ipv6_addr_equal include/net/ipv6.h
From: Dmitry Vyukov @ 2018-06-15 7:32 UTC (permalink / raw)
To: air icy
Cc: David Miller, Alexey Kuznetsov, Hideaki YOSHIFUJI, netdev,
syzkaller
In-Reply-To: <CAAzSK-zy+p3An11LwSRB9Pr9O5oO09u8-XX78QiuLF7j_RFWVg@mail.gmail.com>
On Fri, Jun 15, 2018 at 8:33 AM, air icy <icytxw@gmail.com> wrote:
>
> Hi,
> I found a kernel bug with enchanced syzkaller in the newest linux kernel v4.17.
> The output is as follows:
>
> ==================================================================
> BUG: KASAN: stack-out-of-bounds in ipv6_addr_equal include/net/ipv6.h:508 [inline]
> BUG: KASAN: stack-out-of-bounds in __xfrm6_state_addr_check include/net/xfrm.h:1358 [inline]
> BUG: KASAN: stack-out-of-bounds in xfrm_state_addr_check include/net/xfrm.h:1375 [inline]
> BUG: KASAN: stack-out-of-bounds in xfrm_state_find+0x2693/0x2740 net/xfrm/xfrm_state.c:959
> Read of size 4 at addr ffff880065d77b70 by task syz-executor1/10036
This may be related to "KMSAN: uninit-value in xfrm_state_find":
https://groups.google.com/d/msg/syzkaller-bugs/myqLUHNGRRc/Zb3SlyJZBwAJ
> CPU: 0 PID: 10036 Comm: syz-executor1 Not tainted 4.17.0 #1
> Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.10.2-0-g5f4c7b1-prebuilt.qemu-project.org 04/01/2014
> Call Trace:
>
> The buggy address belongs to the page:
> page:ffffea0001975dc0 count:0 mapcount:0 mapping:0000000000000000 index:0x0
> flags: 0x100000000000000()
> raw: 0100000000000000 0000000000000000 ffffea0001975dc8 0000000000000000
> raw: 0000000000000000 0000000000000000 00000000ffffffff 0000000000000000
> page dumped because: kasan: bad access detected
>
> Memory state around the buggy address:
> ffff880065d77a00: 00 00 00 00 00 00 00 f1 f1 f1 f1 00 f4 f4 f4 f2
> ffff880065d77a80: f2 f2 f2 00 00 00 00 f2 f2 f2 f2 00 00 00 00 00
> >ffff880065d77b00: f4 f4 f4 f2 f2 f2 f2 00 00 00 00 00 00 00 f4 f2
> ^
> ffff880065d77b80: f2 f2 f2 00 00 00 00 00 00 00 00 00 f4 f4 f4 f3
> ffff880065d77c00: f3 f3 f3 00 00 00 00 00 00 00 00 00 00 00 00 00
> ==================================================================
> Kernel panic - not syncing: panic_on_warn set ...
>
> CPU: 0 PID: 10036 Comm: syz-executor1 Tainted: G B 4.17.0 #1
> Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.10.2-0-g5f4c7b1-prebuilt.qemu-project.org 04/01/2014
> Call Trace:
> Dumping ftrace buffer:
> (ftrace buffer empty)
> Kernel Offset: disabled
> Rebooting in 86400 seconds..
>
> bugzilla url: https://bugzilla.kernel.org/show_bug.cgi?id=200065
> config file is attached in this email
>
> thanks
> Xuwen Tu
>
> log2
>
> report2
>
> .config
>
> --
> You received this message because you are subscribed to the Google Groups "syzkaller" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to syzkaller+unsubscribe@googlegroups.com.
> For more options, visit https://groups.google.com/d/optout.
^ permalink raw reply
* Re: KMSAN: uninit-value in xfrm_state_find
From: Dmitry Vyukov @ 2018-06-15 7:31 UTC (permalink / raw)
To: syzbot
Cc: David Miller, Herbert Xu, LKML, netdev, Steffen Klassert,
syzkaller-bugs, icytxw
In-Reply-To: <0000000000001f31eb056ea92fcb@google.com>
On Fri, Jun 15, 2018 at 9:30 AM, syzbot
<syzbot+131cd4c6d21724b99a26@syzkaller.appspotmail.com> wrote:
> Hello,
>
> syzbot found the following crash on:
>
> HEAD commit: 1df165c8d2d6 kmsan: introduce kmsan_clear_user_page()
> git tree: https://github.com/google/kmsan.git/master
> console output: https://syzkaller.appspot.com/x/log.txt?x=15336e97800000
> kernel config: https://syzkaller.appspot.com/x/.config?x=4ca1e57bafa8ab1f
> dashboard link: https://syzkaller.appspot.com/bug?extid=131cd4c6d21724b99a26
> compiler: clang version 7.0.0 (trunk 329391)
> syzkaller repro:https://syzkaller.appspot.com/x/repro.syz?x=12c7a417800000
> C reproducer: https://syzkaller.appspot.com/x/repro.c?x=13710197800000
>
> IMPORTANT: if you fix the bug, please add the following tag to the commit:
> Reported-by: syzbot+131cd4c6d21724b99a26@syzkaller.appspotmail.com
This may be related to "BUG: KASAN: stack-out-of-bounds in
ipv6_addr_equal include/net/ipv6.h":
https://groups.google.com/d/msg/syzkaller/va_9cjZsHQE/Htc7sYY2BwAJ
> IPv6: ADDRCONF(NETDEV_UP): veth1: link is not ready
> IPv6: ADDRCONF(NETDEV_CHANGE): veth1: link becomes ready
> IPv6: ADDRCONF(NETDEV_CHANGE): veth0: link becomes ready
> 8021q: adding VLAN 0 to HW filter on device team0
> ==================================================================
> BUG: KMSAN: uninit-value in __arch_swab32
> arch/x86/include/uapi/asm/swab.h:10 [inline]
> BUG: KMSAN: uninit-value in __fswab32 include/uapi/linux/swab.h:59 [inline]
> BUG: KMSAN: uninit-value in __xfrm6_daddr_saddr_hash net/xfrm/xfrm_hash.h:29
> [inline]
> BUG: KMSAN: uninit-value in __xfrm_dst_hash net/xfrm/xfrm_hash.h:96 [inline]
> BUG: KMSAN: uninit-value in xfrm_dst_hash net/xfrm/xfrm_state.c:60 [inline]
> BUG: KMSAN: uninit-value in xfrm_state_find+0x2b15/0x4f40
> net/xfrm/xfrm_state.c:952
> CPU: 0 PID: 4464 Comm: syz-executor988 Not tainted 4.17.0-rc3+ #93
> Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS
> Google 01/01/2011
> Call Trace:
> __dump_stack lib/dump_stack.c:77 [inline]
> dump_stack+0x185/0x1d0 lib/dump_stack.c:113
> kmsan_report+0x142/0x240 mm/kmsan/kmsan.c:1084
> __msan_warning_32+0x6c/0xb0 mm/kmsan/kmsan_instr.c:683
> __arch_swab32 arch/x86/include/uapi/asm/swab.h:10 [inline]
> __fswab32 include/uapi/linux/swab.h:59 [inline]
> __xfrm6_daddr_saddr_hash net/xfrm/xfrm_hash.h:29 [inline]
> __xfrm_dst_hash net/xfrm/xfrm_hash.h:96 [inline]
> xfrm_dst_hash net/xfrm/xfrm_state.c:60 [inline]
> xfrm_state_find+0x2b15/0x4f40 net/xfrm/xfrm_state.c:952
> xfrm_tmpl_resolve_one net/xfrm/xfrm_policy.c:1393 [inline]
> xfrm_tmpl_resolve net/xfrm/xfrm_policy.c:1437 [inline]
> xfrm_resolve_and_create_bundle+0xc31/0x5270 net/xfrm/xfrm_policy.c:1833
> xfrm_lookup+0x606/0x39d0 net/xfrm/xfrm_policy.c:2163
> xfrm_lookup_route+0xfa/0x360 net/xfrm/xfrm_policy.c:2283
> ip_route_output_flow+0x35b/0x3b0 net/ipv4/route.c:2574
> udp_sendmsg+0x2289/0x33f0 net/ipv4/udp.c:1006
> udpv6_sendmsg+0x1291/0x3f40 net/ipv6/udp.c:1175
> inet_sendmsg+0x48d/0x740 net/ipv4/af_inet.c:798
> sock_sendmsg_nosec net/socket.c:629 [inline]
> sock_sendmsg net/socket.c:639 [inline]
> ___sys_sendmsg+0xec0/0x1310 net/socket.c:2117
> __sys_sendmmsg+0x490/0x850 net/socket.c:2212
> __do_sys_sendmmsg net/socket.c:2241 [inline]
> __se_sys_sendmmsg net/socket.c:2238 [inline]
> __x64_sys_sendmmsg+0x11c/0x170 net/socket.c:2238
> do_syscall_64+0x154/0x220 arch/x86/entry/common.c:287
> entry_SYSCALL_64_after_hwframe+0x44/0xa9
> RIP: 0033:0x4419c9
> RSP: 002b:00007ffdb3fa4608 EFLAGS: 00000217 ORIG_RAX: 0000000000000133
> RAX: ffffffffffffffda RBX: 0000000000000003 RCX: 00000000004419c9
> RDX: 0000000000000001 RSI: 0000000020002000 RDI: 0000000000000003
> RBP: 00000000006cd018 R08: 0000000000000000 R09: 0000000000000000
> R10: 0000000000000000 R11: 0000000000000217 R12: 00000000004026c0
> R13: 0000000000402750 R14: 0000000000000000 R15: 0000000000000000
>
> Local variable description: ----fl4_stack@udp_sendmsg
> Variable was created at:
> udp_sendmsg+0xe5/0x33f0 net/ipv4/udp.c:841
> udpv6_sendmsg+0x1291/0x3f40 net/ipv6/udp.c:1175
> ==================================================================
>
>
> ---
> This bug is generated by a bot. It may contain errors.
> See https://goo.gl/tpsmEJ for more information about syzbot.
> syzbot engineers can be reached at syzkaller@googlegroups.com.
>
> syzbot will keep track of this bug report. See:
> https://goo.gl/tpsmEJ#bug-status-tracking for how to communicate with
> syzbot.
> syzbot can test patches for this bug, for details see:
> https://goo.gl/tpsmEJ#testing-patches
^ permalink raw reply
* KMSAN: uninit-value in xfrm_state_find
From: syzbot @ 2018-06-15 7:30 UTC (permalink / raw)
To: davem, herbert, linux-kernel, netdev, steffen.klassert,
syzkaller-bugs
Hello,
syzbot found the following crash on:
HEAD commit: 1df165c8d2d6 kmsan: introduce kmsan_clear_user_page()
git tree: https://github.com/google/kmsan.git/master
console output: https://syzkaller.appspot.com/x/log.txt?x=15336e97800000
kernel config: https://syzkaller.appspot.com/x/.config?x=4ca1e57bafa8ab1f
dashboard link: https://syzkaller.appspot.com/bug?extid=131cd4c6d21724b99a26
compiler: clang version 7.0.0 (trunk 329391)
syzkaller repro:https://syzkaller.appspot.com/x/repro.syz?x=12c7a417800000
C reproducer: https://syzkaller.appspot.com/x/repro.c?x=13710197800000
IMPORTANT: if you fix the bug, please add the following tag to the commit:
Reported-by: syzbot+131cd4c6d21724b99a26@syzkaller.appspotmail.com
IPv6: ADDRCONF(NETDEV_UP): veth1: link is not ready
IPv6: ADDRCONF(NETDEV_CHANGE): veth1: link becomes ready
IPv6: ADDRCONF(NETDEV_CHANGE): veth0: link becomes ready
8021q: adding VLAN 0 to HW filter on device team0
==================================================================
BUG: KMSAN: uninit-value in __arch_swab32
arch/x86/include/uapi/asm/swab.h:10 [inline]
BUG: KMSAN: uninit-value in __fswab32 include/uapi/linux/swab.h:59 [inline]
BUG: KMSAN: uninit-value in __xfrm6_daddr_saddr_hash
net/xfrm/xfrm_hash.h:29 [inline]
BUG: KMSAN: uninit-value in __xfrm_dst_hash net/xfrm/xfrm_hash.h:96 [inline]
BUG: KMSAN: uninit-value in xfrm_dst_hash net/xfrm/xfrm_state.c:60 [inline]
BUG: KMSAN: uninit-value in xfrm_state_find+0x2b15/0x4f40
net/xfrm/xfrm_state.c:952
CPU: 0 PID: 4464 Comm: syz-executor988 Not tainted 4.17.0-rc3+ #93
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS
Google 01/01/2011
Call Trace:
__dump_stack lib/dump_stack.c:77 [inline]
dump_stack+0x185/0x1d0 lib/dump_stack.c:113
kmsan_report+0x142/0x240 mm/kmsan/kmsan.c:1084
__msan_warning_32+0x6c/0xb0 mm/kmsan/kmsan_instr.c:683
__arch_swab32 arch/x86/include/uapi/asm/swab.h:10 [inline]
__fswab32 include/uapi/linux/swab.h:59 [inline]
__xfrm6_daddr_saddr_hash net/xfrm/xfrm_hash.h:29 [inline]
__xfrm_dst_hash net/xfrm/xfrm_hash.h:96 [inline]
xfrm_dst_hash net/xfrm/xfrm_state.c:60 [inline]
xfrm_state_find+0x2b15/0x4f40 net/xfrm/xfrm_state.c:952
xfrm_tmpl_resolve_one net/xfrm/xfrm_policy.c:1393 [inline]
xfrm_tmpl_resolve net/xfrm/xfrm_policy.c:1437 [inline]
xfrm_resolve_and_create_bundle+0xc31/0x5270 net/xfrm/xfrm_policy.c:1833
xfrm_lookup+0x606/0x39d0 net/xfrm/xfrm_policy.c:2163
xfrm_lookup_route+0xfa/0x360 net/xfrm/xfrm_policy.c:2283
ip_route_output_flow+0x35b/0x3b0 net/ipv4/route.c:2574
udp_sendmsg+0x2289/0x33f0 net/ipv4/udp.c:1006
udpv6_sendmsg+0x1291/0x3f40 net/ipv6/udp.c:1175
inet_sendmsg+0x48d/0x740 net/ipv4/af_inet.c:798
sock_sendmsg_nosec net/socket.c:629 [inline]
sock_sendmsg net/socket.c:639 [inline]
___sys_sendmsg+0xec0/0x1310 net/socket.c:2117
__sys_sendmmsg+0x490/0x850 net/socket.c:2212
__do_sys_sendmmsg net/socket.c:2241 [inline]
__se_sys_sendmmsg net/socket.c:2238 [inline]
__x64_sys_sendmmsg+0x11c/0x170 net/socket.c:2238
do_syscall_64+0x154/0x220 arch/x86/entry/common.c:287
entry_SYSCALL_64_after_hwframe+0x44/0xa9
RIP: 0033:0x4419c9
RSP: 002b:00007ffdb3fa4608 EFLAGS: 00000217 ORIG_RAX: 0000000000000133
RAX: ffffffffffffffda RBX: 0000000000000003 RCX: 00000000004419c9
RDX: 0000000000000001 RSI: 0000000020002000 RDI: 0000000000000003
RBP: 00000000006cd018 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000217 R12: 00000000004026c0
R13: 0000000000402750 R14: 0000000000000000 R15: 0000000000000000
Local variable description: ----fl4_stack@udp_sendmsg
Variable was created at:
udp_sendmsg+0xe5/0x33f0 net/ipv4/udp.c:841
udpv6_sendmsg+0x1291/0x3f40 net/ipv6/udp.c:1175
==================================================================
---
This bug is generated by a bot. It may contain errors.
See https://goo.gl/tpsmEJ for more information about syzbot.
syzbot engineers can be reached at syzkaller@googlegroups.com.
syzbot will keep track of this bug report. See:
https://goo.gl/tpsmEJ#bug-status-tracking for how to communicate with
syzbot.
syzbot can test patches for this bug, for details see:
https://goo.gl/tpsmEJ#testing-patches
^ permalink raw reply
* Re: [PATCH v7 net] stmmac: added support for 802.1ad vlan stripping
From: Toshiaki Makita @ 2018-06-15 7:19 UTC (permalink / raw)
To: Elad Nachman, David Miller
Cc: Jose.Abreu, f.fainelli, netdev, peppe.cavallaro, alexandre.torgue
In-Reply-To: <82c9c00a-b1b7-bd16-0e9c-8b31291cb618@gmail.com>
On 2018/06/15 15:57, Elad Nachman wrote:
> stmmac reception handler calls stmmac_rx_vlan() to strip the vlan before
> calling napi_gro_receive().
>
> The function assumes VLAN tagged frames are always tagged with
> 802.1Q protocol, and assigns ETH_P_8021Q to the skb by hard-coding
> the parameter on call to __vlan_hwaccel_put_tag() .
>
> This causes packets not to be passed to the VLAN slave if it was created
> with 802.1AD protocol
> (ip link add link eth0 eth0.100 type vlan proto 802.1ad id 100).
>
> This fix passes the protocol from the VLAN header into
> __vlan_hwaccel_put_tag() instead of using the hard-coded value of
> ETH_P_8021Q.
>
> NETIF_F_HW_VLAN_STAG_RX check was added and the strip action is now
> dependent on the correct combination of features and the detected vlan tag.
>
> NETIF_F_HW_VLAN_STAG_RX feature was added to be in line with the driver
> actual abilities.
>
> Signed-off-by: Elad Nachman <eladn@gilat.com>
Reviewed-by: Toshiaki Makita <makita.toshiaki@lab.ntt.co.jp>
>
> ---
> drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 21 +++++++++++++--------
> 1 file changed, 13 insertions(+), 8 deletions(-)
>
> diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
> index 11fb7c7..c4ffbfb 100644
> --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
> +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
> @@ -3182,17 +3182,22 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
>
> static void stmmac_rx_vlan(struct net_device *dev, struct sk_buff *skb)
> {
> - struct ethhdr *ehdr;
> + struct vlan_ethhdr *veth;
> + __be16 vlan_proto;
> u16 vlanid;
>
> - if ((dev->features & NETIF_F_HW_VLAN_CTAG_RX) ==
> - NETIF_F_HW_VLAN_CTAG_RX &&
> - !__vlan_get_tag(skb, &vlanid)) {
> + veth = (struct vlan_ethhdr *)skb->data;
> + vlan_proto = veth->h_vlan_proto;
> +
> + if ((vlan_proto == htons(ETH_P_8021Q) &&
> + dev->features & NETIF_F_HW_VLAN_CTAG_RX) ||
> + (vlan_proto == htons(ETH_P_8021AD) &&
> + dev->features & NETIF_F_HW_VLAN_STAG_RX)) {
> /* pop the vlan tag */
> - ehdr = (struct ethhdr *)skb->data;
> - memmove(skb->data + VLAN_HLEN, ehdr, ETH_ALEN * 2);
> + vlanid = ntohs(veth->h_vlan_TCI);
> + memmove(skb->data + VLAN_HLEN, veth, ETH_ALEN * 2);
> skb_pull(skb, VLAN_HLEN);
> - __vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), vlanid);
> + __vlan_hwaccel_put_tag(skb, vlan_proto, vlanid);
> }
> }
>
> @@ -4235,7 +4240,7 @@ int stmmac_dvr_probe(struct device *device,
> ndev->watchdog_timeo = msecs_to_jiffies(watchdog);
> #ifdef STMMAC_VLAN_TAG_USED
> /* Both mac100 and gmac support receive VLAN tag detection */
> - ndev->features |= NETIF_F_HW_VLAN_CTAG_RX;
> + ndev->features |= NETIF_F_HW_VLAN_CTAG_RX | NETIF_F_HW_VLAN_STAG_RX;
> #endif
> priv->msg_enable = netif_msg_init(debug, default_msg_level);
>
>
--
Toshiaki Makita
^ permalink raw reply
* [PATCH v7 net] stmmac: added support for 802.1ad vlan stripping
From: Elad Nachman @ 2018-06-15 6:57 UTC (permalink / raw)
To: Toshiaki Makita, David Miller
Cc: Jose.Abreu, f.fainelli, netdev, peppe.cavallaro, alexandre.torgue,
eladv6
In-Reply-To: <e68da366-9c38-c2d5-80e7-9d5ec8799cdd@lab.ntt.co.jp>
stmmac reception handler calls stmmac_rx_vlan() to strip the vlan before
calling napi_gro_receive().
The function assumes VLAN tagged frames are always tagged with
802.1Q protocol, and assigns ETH_P_8021Q to the skb by hard-coding
the parameter on call to __vlan_hwaccel_put_tag() .
This causes packets not to be passed to the VLAN slave if it was created
with 802.1AD protocol
(ip link add link eth0 eth0.100 type vlan proto 802.1ad id 100).
This fix passes the protocol from the VLAN header into
__vlan_hwaccel_put_tag() instead of using the hard-coded value of
ETH_P_8021Q.
NETIF_F_HW_VLAN_STAG_RX check was added and the strip action is now
dependent on the correct combination of features and the detected vlan tag.
NETIF_F_HW_VLAN_STAG_RX feature was added to be in line with the driver
actual abilities.
Signed-off-by: Elad Nachman <eladn@gilat.com>
---
drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 21 +++++++++++++--------
1 file changed, 13 insertions(+), 8 deletions(-)
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index 11fb7c7..c4ffbfb 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -3182,17 +3182,22 @@ static netdev_tx_t stmmac_xmit(struct sk_buff *skb, struct net_device *dev)
static void stmmac_rx_vlan(struct net_device *dev, struct sk_buff *skb)
{
- struct ethhdr *ehdr;
+ struct vlan_ethhdr *veth;
+ __be16 vlan_proto;
u16 vlanid;
- if ((dev->features & NETIF_F_HW_VLAN_CTAG_RX) ==
- NETIF_F_HW_VLAN_CTAG_RX &&
- !__vlan_get_tag(skb, &vlanid)) {
+ veth = (struct vlan_ethhdr *)skb->data;
+ vlan_proto = veth->h_vlan_proto;
+
+ if ((vlan_proto == htons(ETH_P_8021Q) &&
+ dev->features & NETIF_F_HW_VLAN_CTAG_RX) ||
+ (vlan_proto == htons(ETH_P_8021AD) &&
+ dev->features & NETIF_F_HW_VLAN_STAG_RX)) {
/* pop the vlan tag */
- ehdr = (struct ethhdr *)skb->data;
- memmove(skb->data + VLAN_HLEN, ehdr, ETH_ALEN * 2);
+ vlanid = ntohs(veth->h_vlan_TCI);
+ memmove(skb->data + VLAN_HLEN, veth, ETH_ALEN * 2);
skb_pull(skb, VLAN_HLEN);
- __vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), vlanid);
+ __vlan_hwaccel_put_tag(skb, vlan_proto, vlanid);
}
}
@@ -4235,7 +4240,7 @@ int stmmac_dvr_probe(struct device *device,
ndev->watchdog_timeo = msecs_to_jiffies(watchdog);
#ifdef STMMAC_VLAN_TAG_USED
/* Both mac100 and gmac support receive VLAN tag detection */
- ndev->features |= NETIF_F_HW_VLAN_CTAG_RX;
+ ndev->features |= NETIF_F_HW_VLAN_CTAG_RX | NETIF_F_HW_VLAN_STAG_RX;
#endif
priv->msg_enable = netif_msg_init(debug, default_msg_level);
--
2.7.4
^ permalink raw reply related
* [PATCH RFC ipsec-next] xfrm: Extend the output_mark to support input direction and masking.
From: Steffen Klassert @ 2018-06-15 6:55 UTC (permalink / raw)
To: netdev; +Cc: Tobias Brunner, Eyal Birger, Lorenzo Colitti
We already support setting an output mark at the xfrm_state,
unfortunately this does not support the input direction and
masking the marks that will be applied to the skb. This change
adds support applying a masked value in both directions.
The existing XFRMA_OUTPUT_MARK number is reused for this purpose
and as it is now bi-directional, it is renamed to XFRMA_SET_MARK.
An additional XFRMA_SET_MARK_MASK attribute is added for setting the
mask. If the attribute mask not provided, it is set to 0xffffffff,
keeping the XFRMA_OUTPUT_MARK existing 'full mask' semantics.
Co-developed-by: Tobias Brunner <tobias@strongswan.org>
Co-developed-by: Eyal Birger <eyal.birger@gmail.com>
Co-developed-by: Lorenzo Colitti <lorenzo@google.com>
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
Signed-off-by: Tobias Brunner <tobias@strongswan.org>
Signed-off-by: Eyal Birger <eyal.birger@gmail.com>
Signed-off-by: Lorenzo Colitti <lorenzo@google.com>
---
include/net/xfrm.h | 9 ++++++++-
include/uapi/linux/xfrm.h | 4 +++-
net/xfrm/xfrm_input.c | 2 ++
net/xfrm/xfrm_output.c | 3 +--
net/xfrm/xfrm_policy.c | 5 +++--
net/xfrm/xfrm_user.c | 48 +++++++++++++++++++++++++++++++++++++----------
6 files changed, 55 insertions(+), 16 deletions(-)
diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 45e75c36b738..8727b2484855 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -166,7 +166,7 @@ struct xfrm_state {
int header_len;
int trailer_len;
u32 extra_flags;
- u32 output_mark;
+ struct xfrm_mark smark;
} props;
struct xfrm_lifetime_cfg lft;
@@ -2012,6 +2012,13 @@ static inline int xfrm_mark_put(struct sk_buff *skb, const struct xfrm_mark *m)
return ret;
}
+static inline __u32 xfrm_smark_get(__u32 mark, struct xfrm_state *x)
+{
+ struct xfrm_mark *m = &x->props.smark;
+
+ return (m->v & m->m) | (mark & ~m->m);
+}
+
static inline int xfrm_tunnel_check(struct sk_buff *skb, struct xfrm_state *x,
unsigned int family)
{
diff --git a/include/uapi/linux/xfrm.h b/include/uapi/linux/xfrm.h
index e3af2859188b..5a6ed7ce5a29 100644
--- a/include/uapi/linux/xfrm.h
+++ b/include/uapi/linux/xfrm.h
@@ -305,9 +305,11 @@ enum xfrm_attr_type_t {
XFRMA_ADDRESS_FILTER, /* struct xfrm_address_filter */
XFRMA_PAD,
XFRMA_OFFLOAD_DEV, /* struct xfrm_state_offload */
- XFRMA_OUTPUT_MARK, /* __u32 */
+ XFRMA_SET_MARK, /* __u32 */
+ XFRMA_SET_MARK_MASK, /* __u32 */
__XFRMA_MAX
+#define XFRMA_OUTPUT_MARK XFRMA_SET_MARK /* Compatibility */
#define XFRMA_MAX (__XFRMA_MAX - 1)
};
diff --git a/net/xfrm/xfrm_input.c b/net/xfrm/xfrm_input.c
index 352abca2605f..074810436242 100644
--- a/net/xfrm/xfrm_input.c
+++ b/net/xfrm/xfrm_input.c
@@ -339,6 +339,8 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type)
goto drop;
}
+ skb->mark = xfrm_smark_get(skb->mark, x);
+
skb->sp->xvec[skb->sp->len++] = x;
lock:
diff --git a/net/xfrm/xfrm_output.c b/net/xfrm/xfrm_output.c
index 89b178a78dc7..45ba07ab3e4f 100644
--- a/net/xfrm/xfrm_output.c
+++ b/net/xfrm/xfrm_output.c
@@ -66,8 +66,7 @@ static int xfrm_output_one(struct sk_buff *skb, int err)
goto error_nolock;
}
- if (x->props.output_mark)
- skb->mark = x->props.output_mark;
+ skb->mark = xfrm_smark_get(skb->mark, x);
err = x->outer_mode->output(x, skb);
if (err) {
diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c
index 40b54cc64243..f95f5f75748c 100644
--- a/net/xfrm/xfrm_policy.c
+++ b/net/xfrm/xfrm_policy.c
@@ -1607,10 +1607,11 @@ static struct dst_entry *xfrm_bundle_create(struct xfrm_policy *policy,
dst_copy_metrics(dst1, dst);
if (xfrm[i]->props.mode != XFRM_MODE_TRANSPORT) {
+ __u32 mark = xfrm_smark_get(fl->flowi_mark, xfrm[i]);
+
family = xfrm[i]->props.family;
dst = xfrm_dst_lookup(xfrm[i], tos, fl->flowi_oif,
- &saddr, &daddr, family,
- xfrm[i]->props.output_mark);
+ &saddr, &daddr, family, mark);
err = PTR_ERR(dst);
if (IS_ERR(dst))
goto put_states;
diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c
index 080035f056d9..9602cc9e05ab 100644
--- a/net/xfrm/xfrm_user.c
+++ b/net/xfrm/xfrm_user.c
@@ -527,6 +527,19 @@ static void xfrm_update_ae_params(struct xfrm_state *x, struct nlattr **attrs,
x->replay_maxdiff = nla_get_u32(rt);
}
+static void xfrm_smark_init(struct nlattr **attrs, struct xfrm_mark *m)
+{
+ if (attrs[XFRMA_SET_MARK]) {
+ m->v = nla_get_u32(attrs[XFRMA_SET_MARK]);
+ if (attrs[XFRMA_SET_MARK_MASK])
+ m->m = nla_get_u32(attrs[XFRMA_SET_MARK_MASK]);
+ else
+ m->m = 0xffffffff;
+ } else {
+ m->v = m->m = 0;
+ }
+}
+
static struct xfrm_state *xfrm_state_construct(struct net *net,
struct xfrm_usersa_info *p,
struct nlattr **attrs,
@@ -579,8 +592,7 @@ static struct xfrm_state *xfrm_state_construct(struct net *net,
xfrm_mark_get(attrs, &x->mark);
- if (attrs[XFRMA_OUTPUT_MARK])
- x->props.output_mark = nla_get_u32(attrs[XFRMA_OUTPUT_MARK]);
+ xfrm_smark_init(attrs, &x->props.smark);
err = __xfrm_init_state(x, false, attrs[XFRMA_OFFLOAD_DEV]);
if (err)
@@ -824,6 +836,18 @@ static int copy_to_user_auth(struct xfrm_algo_auth *auth, struct sk_buff *skb)
return 0;
}
+static int xfrm_smark_put(struct sk_buff *skb, struct xfrm_mark *m)
+{
+ int ret = 0;
+
+ if (m->v | m->m) {
+ ret = nla_put_u32(skb, XFRMA_SET_MARK, m->v);
+ if (!ret)
+ ret = nla_put_u32(skb, XFRMA_SET_MARK_MASK, m->m);
+ }
+ return ret;
+}
+
/* Don't change this without updating xfrm_sa_len! */
static int copy_to_user_state_extra(struct xfrm_state *x,
struct xfrm_usersa_info *p,
@@ -887,6 +911,11 @@ static int copy_to_user_state_extra(struct xfrm_state *x,
ret = xfrm_mark_put(skb, &x->mark);
if (ret)
goto out;
+
+ ret = xfrm_smark_put(skb, &x->props.smark);
+ if (ret)
+ goto out;
+
if (x->replay_esn)
ret = nla_put(skb, XFRMA_REPLAY_ESN_VAL,
xfrm_replay_state_esn_len(x->replay_esn),
@@ -900,11 +929,7 @@ static int copy_to_user_state_extra(struct xfrm_state *x,
ret = copy_user_offload(&x->xso, skb);
if (ret)
goto out;
- if (x->props.output_mark) {
- ret = nla_put_u32(skb, XFRMA_OUTPUT_MARK, x->props.output_mark);
- if (ret)
- goto out;
- }
+
if (x->security)
ret = copy_sec_ctx(x->security, skb);
out:
@@ -2493,7 +2518,8 @@ static const struct nla_policy xfrma_policy[XFRMA_MAX+1] = {
[XFRMA_PROTO] = { .type = NLA_U8 },
[XFRMA_ADDRESS_FILTER] = { .len = sizeof(struct xfrm_address_filter) },
[XFRMA_OFFLOAD_DEV] = { .len = sizeof(struct xfrm_user_offload) },
- [XFRMA_OUTPUT_MARK] = { .type = NLA_U32 },
+ [XFRMA_SET_MARK] = { .type = NLA_U32 },
+ [XFRMA_SET_MARK_MASK] = { .type = NLA_U32 },
};
static const struct nla_policy xfrma_spd_policy[XFRMA_SPD_MAX+1] = {
@@ -2719,8 +2745,10 @@ static inline unsigned int xfrm_sa_len(struct xfrm_state *x)
l += nla_total_size(sizeof(x->props.extra_flags));
if (x->xso.dev)
l += nla_total_size(sizeof(x->xso));
- if (x->props.output_mark)
- l += nla_total_size(sizeof(x->props.output_mark));
+ if (x->props.smark.v | x->props.smark.m) {
+ l += nla_total_size(sizeof(x->props.smark.v));
+ l += nla_total_size(sizeof(x->props.smark.m));
+ }
/* Must count x->lastused as it may become non-zero behind our back. */
l += nla_total_size_64bit(sizeof(u64));
--
2.14.1
^ permalink raw reply related
* Re: [PATCH] selftests: bpf: config: add config fragments
From: Anders Roxell @ 2018-06-15 6:41 UTC (permalink / raw)
To: William Tu
Cc: Daniel Borkmann, Alexei Starovoitov, Shuah Khan, Networking,
Linux Kernel Mailing List, open list:KERNEL SELFTEST FRAMEWORK
In-Reply-To: <CALDO+SZXvnUU7VKd_UwnBCU0NiTHjqei4xEUCdcktkKCD3MnEg@mail.gmail.com>
On Thu, 14 Jun 2018 at 14:09, William Tu <u9012063@gmail.com> wrote:
>
> On Thu, Jun 14, 2018 at 4:42 AM, Anders Roxell <anders.roxell@linaro.org> wrote:
> > On 14 June 2018 at 13:06, William Tu <u9012063@gmail.com> wrote:
> >> On Tue, Jun 12, 2018 at 5:08 PM, Daniel Borkmann <daniel@iogearbox.net> wrote:
> >>> On 06/12/2018 01:05 PM, Anders Roxell wrote:
> >>>> Tests test_tunnel.sh fails due to config fragments ins't enabled.
> >>>>
> >>>> Fixes: 933a741e3b82 ("selftests/bpf: bpf tunnel test.")
> >>>> Signed-off-by: Anders Roxell <anders.roxell@linaro.org>
> >>>> ---
> >>>>
> >>>> All tests passes except ip6gretap that still fails. I'm unsure why.
> >>>> Ideas?
> >>
> >> Hi Anders,
> >>
> >> ip6erspan is based on ip6gretap, does ip6erspan pass?
> >
> > it did pass when I was sending the email.
> > However, I retested this on next-20180613 and now it fails.
> >
> Does 'ip -s link show' show any errors/dropped on ip6gretap device?
I rerun the test_ip6gretap test only and added "set -x" to
test_tunnel.sh here's the output.
I added "ip -s link show ip6gretap11" before the cleanup function in the script.
# ./test_tunnel.sh
+ PING_ARG='-c 3 -w 10 -q'
+ ret=0
+ GREEN='\033[0;92m'
+ RED='\033[0;31m'
+ NC='\033[0m'
+ trap cleanup 0 3 6
+ trap cleanup_exit 2 9
+ cleanup
+ ip netns delete at_ns0
+ ip link del veth1
+ ip link del ipip11
+ ip link del ipip6tnl11
+ ip link del gretap11
+ ip link del ip6gre11
+ ip link del ip6gretap11
+ ip link del vxlan11
+ ip link del ip6vxlan11
+ ip link del geneve11
+ ip link del ip6geneve11
+ ip link del erspan11
+ ip link del ip6erspan11
+ bpf_tunnel_test
+ echo 'Testing IP6GRETAP tunnel...'
Testing IP6GRETAP tunnel...
+ test_ip6gretap
+ TYPE=ip6gretap
+ DEV_NS=ip6gretap00
+ DEV=ip6gretap11
+ ret=0
+ check ip6gretap
+ ip link help ip6gretap
+ grep -q '^Usage:'
+ '[' 0 -ne 0 ']'
+ config_device
+ ip netns add at_ns0
+ ip link add veth0 type veth peer name veth1
+ ip link set veth0 netns at_ns0
+ ip netns exec at_ns0 ip addr add 172.16.1.100/24 dev veth0
+ ip netns exec at_ns0 ip link set dev veth0 up
+ ip link set dev veth1 up mtu 1500
+ ip addr add dev veth1 172.16.1.200/24
+ add_ip6gretap_tunnel
+ ip netns exec at_ns0 ip addr add ::11/96 dev veth0
+ ip netns exec at_ns0 ip link set dev veth0 up
+ ip addr add dev veth1 ::22/96
+ ip link set dev veth1 up
+ ip netns exec at_ns0 ip link add dev ip6gretap00 type ip6gretap seq
flowlabel 0xbcdef key 2 local ::11 remote ::22
+ ip netns exec at_ns0 ip addr add dev ip6gretap00 10.1.1.100/24
+ ip netns exec at_ns0 ip addr add dev ip6gretap00 fc80::100/96
+ ip netns exec at_ns0 ip link set dev ip6gretap00 up
+ ip link add dev ip6gretap11 type ip6gretap external
+ ip addr add dev ip6gretap11 10.1.1.200/24
+ ip addr add dev ip6gretap11 fc80::200/24
+ ip link set dev ip6gretap11 up
+ attach_bpf ip6gretap11 ip6gretap_set_tunnel ip6gretap_get_tunnel
+ DEV=ip6gretap11
+ SET=ip6gretap_set_tunnel
+ GET=ip6gretap_get_tunnel
+ tc qdisc add dev ip6gretap11 clsact
+ tc filter add dev ip6gretap11 egress bpf da obj test_tunnel_kern.o
sec ip6gretap_set_tunnel
+ tc filter add dev ip6gretap11 ingress bpf da obj test_tunnel_kern.o
sec ip6gretap_get_tunnel
+ ping6 -c 3 -w 10 -q ::11
PING ::11 (::11): 56 data bytes
--- ::11 ping statistics ---
5 packets transmitted, 3 packets received, 40% packet loss
round-trip min/avg/max = 0.139/1.857/5.293 ms
+ ip netns exec at_ns0 ping -c 3 -w 10 -q 10.1.1.200
PING 10.1.1.200 (10.1.1.200): 56 data bytes
--- 10.1.1.200 ping statistics ---
3 packets transmitted, 3 packets received, 0% packet loss
round-trip min/avg/max = 0.214/0.256/0.305 ms
+ ping -c 3 -w 10 -q 10.1.1.100
PING 10.1.1.100 (10.1.1.100): 56 data bytes
--- 10.1.1.100 ping statistics ---
3 packets transmitted, 3 packets received, 0% packet loss
round-trip min/avg/max = 0.210/0.211/0.213 ms
+ check_err 0
+ '[' 0 -eq 0 ']'
+ ret=0
+ ip netns exec at_ns0 ping6 -c 3 -w 10 -q fc80::200
PING fc80::200 (fc80::200): 56 data bytes
--- fc80::200 ping statistics ---
10 packets transmitted, 0 packets received, 100% packet loss
+ check_err 1
+ '[' 0 -eq 0 ']'
+ ret=1
+ ip -s link show ip6gretap11
19: ip6gretap11@NONE: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1434 qdisc
pfifo_fast state UNKNOWN mode DEFAULT group default qlen 1000
link/ether de:d2:0c:53:80:8c brd ff:ff:ff:ff:ff:ff
RX: bytes packets errors dropped overrun mcast
2096 25 0 0 0 0
TX: bytes packets errors dropped carrier collsns
5324 36 5 5 0 0
+ cleanup
+ ip netns delete at_ns0
+ ip link del veth1
+ ip link del ipip11
+ ip link del ipip6tnl11
+ ip link del gretap11
+ ip link del ip6gre11
+ ip link del ip6gretap11
+ ip link del vxlan11
+ ip link del ip6vxlan11
+ ip link del geneve11
+ ip link del ip6geneve11
+ ip link del erspan11
+ ip link del ip6erspan11
+ '[' 1 -ne 0 ']'
+ echo -e '\033[0;31mFAIL: ip6gretap\033[0m'
FAIL: ip6gretap
+ return 1
+ exit 0
+ cleanup
+ ip netns delete at_ns0
+ ip link del veth1
+ ip link del ipip11
+ ip link del ipip6tnl11
+ ip link del gretap11
+ ip link del ip6gre11
+ ip link del ip6gretap11
+ ip link del vxlan11
+ ip link del ip6vxlan11
+ ip link del geneve11
+ ip link del ip6geneve11
+ ip link del erspan11
+ ip link del ip6erspan11
^ permalink raw reply
* Re: [PATCH net-next,RFC 00/13] New fast forwarding path
From: Steffen Klassert @ 2018-06-15 6:34 UTC (permalink / raw)
To: David Miller; +Cc: tom, pablo, netfilter-devel, netdev
In-Reply-To: <20180614.165834.338565136334574983.davem@davemloft.net>
On Thu, Jun 14, 2018 at 04:58:34PM -0700, David Miller wrote:
> From: Tom Herbert <tom@herbertland.com>
> Date: Thu, 14 Jun 2018 13:52:03 -0700
>
> > IIRC, there was a similar proposal a while back that want to bundle
> > packets of the same flow together (without doing GRO) so that they
> > could be processed by various functions by looking at just one
> > representative packet in the group. The concept had some promise, but
> > in the end it created quite a bit of complexity since at some point
> > the packet bundle needed to be undone to go back to processing the
> > individual packets.
>
> You're probably talking about Edward Cree's SKB list stuff, and as
> per his presenation at netconf 2 weeks ago he plans to revitalize
> it given how Spectre et al. gives cause to reevaluate all bulking
> techniques.
Are there patches for the proposal Edward did a while ago,
or was it just a concept?
Maybe we can somehow put things together, I just need some
batching method that works for IPsec and UDP. It does not
need to be exactly the one we proposing here.
^ permalink raw reply
* Re: [PATCH net-next,RFC 00/13] New fast forwarding path
From: Steffen Klassert @ 2018-06-15 6:27 UTC (permalink / raw)
To: Tom Herbert
Cc: Pablo Neira Ayuso, netfilter-devel,
Linux Kernel Network Developers
In-Reply-To: <CALx6S345B2-sejH52beewdiHuGo5nAY0v0uFmNuiJ0hQ6Wu3xw@mail.gmail.com>
On Thu, Jun 14, 2018 at 01:52:03PM -0700, Tom Herbert wrote:
> On Thu, Jun 14, 2018 at 7:19 AM, Pablo Neira Ayuso <pablo@netfilter.org> wrote:
> > Hi,
> >
> > This patchset proposes a new fast forwarding path infrastructure that
> > combines the GRO/GSO and the flowtable infrastructures. The idea is to
> > add a hook at the GRO layer that is invoked before the standard GRO
> > protocol offloads. This allows us to build custom packet chains that we
> > can quickly pass in one go to the neighbour layer to define fast
> > forwarding path for flows.
> >
> > For each packet that gets into the GRO layer, we first check if there is
> > an entry in the flowtable, if so, the packet is placed in a list until
> > the GRO infrastructure decides to send the batch from gro_complete to
> > the neighbour layer. The first packet in the list takes the route from
> > the flowtable entry, so we avoid reiterative routing lookups.
> >
> > In case no entry is found in the flowtable, the packet is passed up to
> > the classic GRO offload handlers. Thus, this packet follows the standard
> > forwarding path. Note that the initial packets of the flow always go
> > through the standard IPv4/IPv6 netfilter forward hook, that is used to
> > configure what flows are placed in the flowtable. Therefore, only a few
> > (initial) packets follow the standard forwarding path while most of the
> > follow up packets take this new fast forwarding path.
> >
>
> IIRC, there was a similar proposal a while back that want to bundle
> packets of the same flow together (without doing GRO) so that they
> could be processed by various functions by looking at just one
> representative packet in the group. The concept had some promise, but
> in the end it created quite a bit of complexity since at some point
> the packet bundle needed to be undone to go back to processing the
> individual packets.
With the way we chain the packets it is not too complicated to
undo this chaining (nft_skb_segment in patch 5 implements this).
After that, this looks like a chain of usual segments, so we
trigger xmit_more with every packet chain.
^ permalink raw reply
* Re: [PATCH net-next,RFC 00/13] New fast forwarding path
From: Steffen Klassert @ 2018-06-15 6:17 UTC (permalink / raw)
To: David Miller; +Cc: pablo, netfilter-devel, netdev
In-Reply-To: <20180614.101831.465275975690050595.davem@davemloft.net>
On Thu, Jun 14, 2018 at 10:18:31AM -0700, David Miller wrote:
> From: Pablo Neira Ayuso <pablo@netfilter.org>
> Date: Thu, 14 Jun 2018 16:19:34 +0200
>
> > This patchset proposes a new fast forwarding path infrastructure
> > that combines the GRO/GSO and the flowtable infrastructures. The
> > idea is to add a hook at the GRO layer that is invoked before the
> > standard GRO protocol offloads. This allows us to build custom
> > packet chains that we can quickly pass in one go to the neighbour
> > layer to define fast forwarding path for flows.
>
> We have full, complete, customizability of the packet path via XDP
> and eBPF.
>
> XDP and eBPF supports everything necessary to accomplish that,
> there are implementations of forwarding implementations in
> the tree and elsewhere.
>
> And most importantly, XDP and eBPF are optimized in drivers and
> offloaded to hardware.
>
> There really is no need for something like what you are proposing.
I started with this last year because I wanted to improve
the IPsec (and UDP) forwarding path. Batching packets
at layer2 and send them directly to the output path
seemed to be a good method to improve this.
In particular, we need to do only one IPsec lookup
for the whole packet chain. So it relaxes the pain
from reomoving the IPsec flowcache a bit. It can be
only a first step, but we need some improvements here
as people start to complain about that.
^ permalink raw reply
* Re: [bpf PATCH v2 6/6] bpf: selftest remove attempts to add LISTEN sockets to sockmap
From: Martin KaFai Lau @ 2018-06-15 6:07 UTC (permalink / raw)
To: John Fastabend; +Cc: ast, daniel, netdev
In-Reply-To: <20180614164512.24994.56879.stgit@john-Precision-Tower-5810>
On Thu, Jun 14, 2018 at 09:45:12AM -0700, John Fastabend wrote:
> In selftest test_maps the sockmap test case attempts to add a socket
> in listening state to the sockmap. This is no longer a valid operation
> so it fails as expected. However, the test wrongly reports this as an
> error now. Fix the test to avoid adding sockets in listening state.
>
> Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Acked-by: Martin KaFai Lau <kafai@fb.com>
> ---
> 0 files changed
>
> diff --git a/tools/testing/selftests/bpf/test_maps.c b/tools/testing/selftests/bpf/test_maps.c
> index 6c25334..9fed5f0 100644
> --- a/tools/testing/selftests/bpf/test_maps.c
> +++ b/tools/testing/selftests/bpf/test_maps.c
> @@ -564,7 +564,7 @@ static void test_sockmap(int tasks, void *data)
> }
>
> /* Test update without programs */
> - for (i = 0; i < 6; i++) {
> + for (i = 2; i < 6; i++) {
> err = bpf_map_update_elem(fd, &i, &sfd[i], BPF_ANY);
> if (err) {
> printf("Failed noprog update sockmap '%i:%i'\n",
> @@ -727,7 +727,7 @@ static void test_sockmap(int tasks, void *data)
> }
>
> /* Test map update elem afterwards fd lives in fd and map_fd */
> - for (i = 0; i < 6; i++) {
> + for (i = 2; i < 6; i++) {
> err = bpf_map_update_elem(map_fd_rx, &i, &sfd[i], BPF_ANY);
> if (err) {
> printf("Failed map_fd_rx update sockmap %i '%i:%i'\n",
>
^ permalink raw reply
* Re: [bpf PATCH v2 5/6] bpf: sockhash, add release routine
From: Martin KaFai Lau @ 2018-06-15 6:05 UTC (permalink / raw)
To: John Fastabend; +Cc: ast, daniel, netdev
In-Reply-To: <20180614164507.24994.83616.stgit@john-Precision-Tower-5810>
On Thu, Jun 14, 2018 at 09:45:07AM -0700, John Fastabend wrote:
> Add map_release_uref pointer to hashmap ops. This was dropped when
> original sockhash code was ported into bpf-next before initial
> commit.
>
> Fixes: 81110384441a ("bpf: sockmap, add hash map support")
> Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Acked-by: Martin KaFai Lau <kafai@fb.com>
> ---
> 0 files changed
>
> diff --git a/kernel/bpf/sockmap.c b/kernel/bpf/sockmap.c
> index ffc5152..77fe204 100644
> --- a/kernel/bpf/sockmap.c
> +++ b/kernel/bpf/sockmap.c
> @@ -2518,6 +2518,7 @@ struct sock *__sock_hash_lookup_elem(struct bpf_map *map, void *key)
> .map_get_next_key = sock_hash_get_next_key,
> .map_update_elem = sock_hash_update_elem,
> .map_delete_elem = sock_hash_delete_elem,
> + .map_release_uref = sock_map_release,
> };
>
> static bool bpf_is_valid_sock(struct bpf_sock_ops_kern *ops)
>
^ permalink raw reply
* Re: [bpf PATCH v2 4/6] bpf: sockmap, tcp_disconnect to listen transition
From: Martin KaFai Lau @ 2018-06-15 6:04 UTC (permalink / raw)
To: John Fastabend; +Cc: ast, daniel, netdev
In-Reply-To: <20180614164502.24994.38682.stgit@john-Precision-Tower-5810>
On Thu, Jun 14, 2018 at 09:45:02AM -0700, John Fastabend wrote:
> After adding checks to ensure TCP is in ESTABLISHED state when a
> sock is added we need to also ensure that user does not transition
> through tcp_disconnect() and back into ESTABLISHED state without
> sockmap removing the sock.
>
> To do this add unhash hook and remove sock from map there.
>
> Reported-by: Eric Dumazet <edumazet@google.com>
> Fixes: 81110384441a ("bpf: sockmap, add hash map support")
> Signed-off-by: John Fastabend <john.fastabend@gmail.com>
LGTM. One nit.
Acked-by: Martin KaFai Lau <kafai@fb.com>
> ---
> 0 files changed
>
> diff --git a/kernel/bpf/sockmap.c b/kernel/bpf/sockmap.c
> index 04764f5..ffc5152 100644
> --- a/kernel/bpf/sockmap.c
> +++ b/kernel/bpf/sockmap.c
> @@ -130,6 +130,7 @@ struct smap_psock {
>
> struct proto *sk_proto;
> void (*save_close)(struct sock *sk, long timeout);
> + void (*save_unhash)(struct sock *sk);
> void (*save_data_ready)(struct sock *sk);
> void (*save_write_space)(struct sock *sk);
> };
> @@ -141,6 +142,7 @@ static int bpf_tcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len,
> static int bpf_tcp_sendpage(struct sock *sk, struct page *page,
> int offset, size_t size, int flags);
> static void bpf_tcp_close(struct sock *sk, long timeout);
> +static void bpf_tcp_unhash(struct sock *sk);
>
> static inline struct smap_psock *smap_psock_sk(const struct sock *sk)
> {
> @@ -182,6 +184,7 @@ static void build_protos(struct proto prot[SOCKMAP_NUM_CONFIGS],
> {
> prot[SOCKMAP_BASE] = *base;
> prot[SOCKMAP_BASE].close = bpf_tcp_close;
> + prot[SOCKMAP_BASE].unhash = bpf_tcp_unhash;
> prot[SOCKMAP_BASE].recvmsg = bpf_tcp_recvmsg;
> prot[SOCKMAP_BASE].stream_memory_read = bpf_tcp_stream_read;
>
> @@ -215,6 +218,7 @@ static int bpf_tcp_init(struct sock *sk)
> }
>
> psock->save_close = sk->sk_prot->close;
> + psock->save_unhash = sk->sk_prot->unhash;
> psock->sk_proto = sk->sk_prot;
>
> /* Build IPv6 sockmap whenever the address of tcpv6_prot changes */
> @@ -302,28 +306,12 @@ struct smap_psock_map_entry *psock_map_pop(struct sock *sk,
> return e;
> }
>
> -static void bpf_tcp_close(struct sock *sk, long timeout)
> +static void bpf_tcp_remove(struct sock *sk, struct smap_psock *psock)
> {
> - void (*close_fun)(struct sock *sk, long timeout);
> struct smap_psock_map_entry *e;
> struct sk_msg_buff *md, *mtmp;
> - struct smap_psock *psock;
> struct sock *osk;
>
> - rcu_read_lock();
> - psock = smap_psock_sk(sk);
> - if (unlikely(!psock)) {
> - rcu_read_unlock();
> - return sk->sk_prot->close(sk, timeout);
> - }
> -
> - /* The psock may be destroyed anytime after exiting the RCU critial
> - * section so by the time we use close_fun the psock may no longer
> - * be valid. However, bpf_tcp_close is called with the sock lock
> - * held so the close hook and sk are still valid.
> - */
> - close_fun = psock->save_close;
> -
> if (psock->cork) {
> free_start_sg(psock->sock, psock->cork);
> kfree(psock->cork);
> @@ -378,6 +366,51 @@ static void bpf_tcp_close(struct sock *sk, long timeout)
> }
> e = psock_map_pop(sk, psock);
> }
> +}
> +
> +static void bpf_tcp_unhash(struct sock *sk)
> +{
> + void (*unhash_fun)(struct sock *sk);
> + struct smap_psock *psock;
> +
> + rcu_read_lock();
> + psock = smap_psock_sk(sk);
> + if (unlikely(!psock)) {
> + rcu_read_unlock();
> + return sk->sk_prot->unhash(sk);
> + }
> +
> + /* The psock may be destroyed anytime after exiting the RCU critial
> + * section so by the time we use close_fun the psock may no longer
> + * be valid. However, bpf_tcp_close is called with the sock lock
> + * held so the close hook and sk are still valid.
> + */
Nit. s/close/unhash/
> + unhash_fun = psock->save_unhash;
> + bpf_tcp_remove(sk, psock);
> + rcu_read_unlock();
> + unhash_fun(sk);
> +
> +}
> +
> +static void bpf_tcp_close(struct sock *sk, long timeout)
> +{
> + void (*close_fun)(struct sock *sk, long timeout);
> + struct smap_psock *psock;
> +
> + rcu_read_lock();
> + psock = smap_psock_sk(sk);
> + if (unlikely(!psock)) {
> + rcu_read_unlock();
> + return sk->sk_prot->close(sk, timeout);
> + }
> +
> + /* The psock may be destroyed anytime after exiting the RCU critial
> + * section so by the time we use close_fun the psock may no longer
> + * be valid. However, bpf_tcp_close is called with the sock lock
> + * held so the close hook and sk are still valid.
> + */
> + close_fun = psock->save_close;
> + bpf_tcp_remove(sk, psock);
> rcu_read_unlock();
> close_fun(sk, timeout);
> }
>
^ permalink raw reply
* Re: [PATCH net-next,RFC 00/13] New fast forwarding path
From: Steffen Klassert @ 2018-06-15 6:03 UTC (permalink / raw)
To: Eric Dumazet; +Cc: Pablo Neira Ayuso, netfilter-devel, netdev
In-Reply-To: <70e4a28c-ccc3-6fd5-0d43-08ae72b7ad1b@gmail.com>
On Thu, Jun 14, 2018 at 08:57:20AM -0700, Eric Dumazet wrote:
>
>
> On 06/14/2018 07:19 AM, Pablo Neira Ayuso wrote:
> > Hi,
> >
>
> > We have collected performance numbers:
> >
> > TCP TSO TCP Fast Forward
> > 32.5 Gbps 35.6 Gbps
> >
> > UDP UDP Fast Forward
> > 17.6 Gbps 35.6 Gbps
> >
> > ESP ESP Fast Forward
> > 6 Gbps 7.5 Gbps
> >
> > For UDP, this is doubling performance, and we almost achieve line rate
> > with one single CPU using the Intel i40e NIC. We got similar numbers
> > with the Mellanox ConnectX-4. For TCP, this is slightly improving things
> > even if TSO is being defeated given that we need to segment the packet
> > chain in software. We would like to explore HW GRO support with hardware
> > vendors with this new mode, we think that should improve the TCP numbers
> > we are showing above even more.
>
> Hi Pablo
>
> Not very convincing numbers, because it is unclear what traffic patterns were used.
>
> We normally use packets per second to measure a forwarding workload,
> and it is not clear if you tried a DDOS, or/and a mix of packets being locally
> delivered and packets being forwarded.
Yes, these number need some more explaination. We used my IPsec
forwarding test setup for this. It looks like this:
------------ ------------
-->| router 1 |-------->| router 2 |--
| ------------ ------------ |
| |
| -------------------- |
--------|Spirent Testcenter|<----------
--------------------
The numbers are from single stream forwarding tests, no local delivery.
Packet size in the UDP case was 1460 byte. I used this packet size
because such packets still fit into the mtu when encapsulated by IPsec.
>
> Presumably adding cache line misses (to probe for flows) will slow down the things.
>
> I suspect the NIC you use has some kind of bottleneck on sending TSO packets,
> or that you hit the issue that GRO might cook suboptimal packets for forwarding workloads
> (eg setting frag_list)
That might be, I was a bit surprised about the TCP numbers myself.
I was more focused on UDP and IPsec because these don't have
hardware segmentation support. I've just added a TCP handler to
see what happens, the numbers looked ok, so I kept it.
All this is based on the approach I pesented last year at the nefilter
workshop.
>
> This path series add yet more code to GRO engine which is already very fat
> to the point many people advocate to turn it off.
We tried to stay away from the generic codepath as much as possible.
Currently we need five 'if' statements, two of them are in error
paths (Patch 4).
> Saving cpu cycles on moderate load is not okay if added complexity
> slows down the DDOS (or stress) by 10 % :/
Why 10%?
>
> To me, GRO is specialized to optimize the non-forwarding case,
> so it is counter-intuitive to base a fast forwarding path on top of it.
It is optimized for the non-forwarding case, but it seems that forwarding
can benefit from that too with very little cost for the non-forwarding case.
^ permalink raw reply
* Re: [bpf PATCH v2 3/6] bpf: sockhash fix omitted bucket lock in sock_close
From: Martin KaFai Lau @ 2018-06-15 5:41 UTC (permalink / raw)
To: John Fastabend; +Cc: ast, daniel, netdev
In-Reply-To: <20180614164457.24994.33710.stgit@john-Precision-Tower-5810>
On Thu, Jun 14, 2018 at 09:44:57AM -0700, John Fastabend wrote:
> First in tcp_close, reduce scope of sk_callback_lock() the lock is
> only needed for protecting smap_release_sock() the ingress and cork
> lists are protected by sock lock. Having the lock in wider scope is
> harmless but may confuse the reader who may infer it is in fact
> needed.
>
> Next, in sock_hash_delete_elem() the pattern is as follows,
>
> sock_hash_delete_elem()
> [...]
> spin_lock(bucket_lock)
> l = lookup_elem_raw()
> if (l)
> hlist_del_rcu()
> write_lock(sk_callback_lock)
> .... destroy psock ...
> write_unlock(sk_callback_lock)
> spin_unlock(bucket_lock)
>
> The ordering is necessary because we only know the {p}sock after
> dereferencing the hash table which we can't do unless we have the
> bucket lock held. Once we have the bucket lock and the psock element
> it is deleted from the hashmap to ensure any other path doing a lookup
> will fail. Finally, the refcnt is decremented and if zero the psock
> is destroyed.
>
> In parallel with the above (or free'ing the map) a tcp close event
> may trigger tcp_close(). Which at the moment omits the bucket lock
> altogether (oops!) where the flow looks like this,
>
> bpf_tcp_close()
> [...]
> write_lock(sk_callback_lock)
> for each psock->maps // list of maps this sock is part of
> hlist_del_rcu(ref_hash_node);
> .... destroy psock ...
> write_unlock(sk_callback_lock)
>
> Obviously, and demonstrated by syzbot, this is broken because
> we can have multiple threads deleting entries via hlist_del_rcu().
>
> To fix this we might be tempted to wrap the hlist operation in a
> bucket lock but that would create a lock inversion problem. In
> summary to follow locking rules maps needs the sk_callback_lock but we
> need the bucket lock to do the hlist_del_rcu. To resolve the lock
> inversion problem note that when bpf_tcp_close is called no updates
> can happen in parallel, due to ESTABLISH state check in update logic,
> so pop the head of the list repeatedly and remove the reference until
> no more are left. If a delete happens in parallel from the BPF API
> that is OK as well because it will do a similar action, lookup the
> sock in the map/hash, delete it from the map/hash, and dec the refcnt.
> We check for this case before doing a destroy on the psock to ensure
> we don't have two threads tearing down a psock. The new logic is
> as follows,
>
> bpf_tcp_close()
> e = psock_map_pop(psock->maps) // done with sk_callback_lock
> bucket_lock() // lock hash list bucket
> l = lookup_elem_raw(head, hash, key, key_size);
> if (l) {
> //only get here if elmnt was not already removed
> hlist_del_rcu()
> ... destroy psock...
> }
> bucket_unlock()
>
> And finally for all the above to work add missing sk_callback_lock
> around smap_list_remove in sock_hash_ctx_update_elem(). Otherwise
> delete and update may corrupt maps list.
>
> (As an aside the sk_callback_lock serves two purposes. The
> first, is to update the sock callbacks sk_data_ready, sk_write_space,
> etc. The second is to protect the psock 'maps' list. The 'maps' list
> is used to (as shown above) to delete all map/hash references to a
> sock when the sock is closed)
>
> (If we did not have the ESTABLISHED state guarantee from tcp_close
> then we could not ensure completion because updates could happen
> forever and pin thread in delete loop.)
>
> Reported-by: syzbot+0ce137753c78f7b6acc1@syzkaller.appspotmail.com
> Fixes: 81110384441a ("bpf: sockmap, add hash map support")
> Signed-off-by: John Fastabend <john.fastabend@gmail.com>
> ---
> 0 files changed
>
> diff --git a/kernel/bpf/sockmap.c b/kernel/bpf/sockmap.c
> index f1ab52d..04764f5 100644
> --- a/kernel/bpf/sockmap.c
> +++ b/kernel/bpf/sockmap.c
> @@ -258,16 +258,54 @@ static void bpf_tcp_release(struct sock *sk)
> rcu_read_unlock();
> }
>
> +static struct htab_elem *lookup_elem_raw(struct hlist_head *head,
> + u32 hash, void *key, u32 key_size)
> +{
> + struct htab_elem *l;
> +
> + hlist_for_each_entry_rcu(l, head, hash_node) {
> + if (l->hash == hash && !memcmp(&l->key, key, key_size))
> + return l;
> + }
> +
> + return NULL;
> +}
> +
> +static inline struct bucket *__select_bucket(struct bpf_htab *htab, u32 hash)
> +{
> + return &htab->buckets[hash & (htab->n_buckets - 1)];
> +}
> +
> +static inline struct hlist_head *select_bucket(struct bpf_htab *htab, u32 hash)
> +{
> + return &__select_bucket(htab, hash)->head;
> +}
> +
> static void free_htab_elem(struct bpf_htab *htab, struct htab_elem *l)
> {
> atomic_dec(&htab->count);
> kfree_rcu(l, rcu);
> }
>
> +struct smap_psock_map_entry *psock_map_pop(struct sock *sk,
> + struct smap_psock *psock)
> +{
> + struct smap_psock_map_entry *e;
> +
> + write_lock_bh(&sk->sk_callback_lock);
> + e = list_first_entry_or_null(&psock->maps,
> + struct smap_psock_map_entry,
> + list);
> + if (e)
> + list_del(&e->list);
> + write_unlock_bh(&sk->sk_callback_lock);
> + return e;
> +}
> +
> static void bpf_tcp_close(struct sock *sk, long timeout)
> {
> void (*close_fun)(struct sock *sk, long timeout);
> - struct smap_psock_map_entry *e, *tmp;
> + struct smap_psock_map_entry *e;
> struct sk_msg_buff *md, *mtmp;
> struct smap_psock *psock;
> struct sock *osk;
> @@ -286,7 +324,6 @@ static void bpf_tcp_close(struct sock *sk, long timeout)
> */
> close_fun = psock->save_close;
>
> - write_lock_bh(&sk->sk_callback_lock);
> if (psock->cork) {
> free_start_sg(psock->sock, psock->cork);
> kfree(psock->cork);
> @@ -299,20 +336,48 @@ static void bpf_tcp_close(struct sock *sk, long timeout)
> kfree(md);
> }
>
> - list_for_each_entry_safe(e, tmp, &psock->maps, list) {
> + /* Sock is in TCP_CLOSE state so any concurrent adds or updates will be
> + * blocked by ESTABLISHED check. However, tcp_close() + delete + free
> + * can all run at the same time. If a tcp_close + delete happens each
> + * code path will remove the entry for the map/hash before deleting it.
> + * In the map case a xchg and then check to verify we have a sk protects
> + * two paths from tearing down the same object. For hash map we lock the
> + * bucket and remove the object from the hash map before destroying to
> + * ensure that only one reference exists. By pulling object off the head
> + * of the list with (with sk_callback_lock) if multiple deleters are
> + * running we avoid duplicate references.
> + */
> + e = psock_map_pop(sk, psock);
> + while (e) {
> if (e->entry) {
> osk = cmpxchg(e->entry, sk, NULL);
> if (osk == sk) {
> - list_del(&e->list);
> smap_release_sock(psock, sk);
> }
> } else {
> - hlist_del_rcu(&e->hash_link->hash_node);
> - smap_release_sock(psock, e->hash_link->sk);
> - free_htab_elem(e->htab, e->hash_link);
> + struct htab_elem *link = e->hash_link;
> + struct hlist_head *head;
> + struct htab_elem *l;
> + struct bucket *b;
> +
> + b = __select_bucket(e->htab, link->hash);
> + head = &b->head;
> + raw_spin_lock_bh(&b->lock);
> + l = lookup_elem_raw(head,
> + link->hash, link->key,
> + e->htab->elem_size);
> + /* If another thread deleted this object skip deletion.
> + * The refcnt on psock may or may not be zero.
> + */
> + if (l) {
> + hlist_del_rcu(&e->hash_link->hash_node);
> + smap_release_sock(psock, e->hash_link->sk);
> + free_htab_elem(e->htab, e->hash_link);
> + }
> + raw_spin_unlock_bh(&b->lock);
> }
> + e = psock_map_pop(sk, psock);
> }
> - write_unlock_bh(&sk->sk_callback_lock);
> rcu_read_unlock();
> close_fun(sk, timeout);
> }
> @@ -2088,16 +2153,6 @@ static struct bpf_map *sock_hash_alloc(union bpf_attr *attr)
> return ERR_PTR(err);
> }
>
> -static inline struct bucket *__select_bucket(struct bpf_htab *htab, u32 hash)
> -{
> - return &htab->buckets[hash & (htab->n_buckets - 1)];
> -}
> -
> -static inline struct hlist_head *select_bucket(struct bpf_htab *htab, u32 hash)
> -{
> - return &__select_bucket(htab, hash)->head;
> -}
> -
> static void sock_hash_free(struct bpf_map *map)
> {
> struct bpf_htab *htab = container_of(map, struct bpf_htab, map);
> @@ -2114,10 +2169,13 @@ static void sock_hash_free(struct bpf_map *map)
> */
> rcu_read_lock();
> for (i = 0; i < htab->n_buckets; i++) {
> - struct hlist_head *head = select_bucket(htab, i);
> + struct bucket *b = __select_bucket(htab, i);
> + struct hlist_head *head;
> struct hlist_node *n;
> struct htab_elem *l;
>
> + raw_spin_lock_bh(&b->lock);
There is a synchronize_rcu() at the beginning of sock_hash_free().
Taking the bucket's lock of the free-ing map at this point is a bit
suspicious. What may still access the map after synchronize_rcu()?
> + head = &b->head;
> hlist_for_each_entry_safe(l, n, head, hash_node) {
> struct sock *sock = l->sk;
> struct smap_psock *psock;
> @@ -2137,6 +2195,7 @@ static void sock_hash_free(struct bpf_map *map)
> write_unlock_bh(&sock->sk_callback_lock);
> kfree(l);
> }
> + raw_spin_unlock_bh(&b->lock);
> }
> rcu_read_unlock();
> bpf_map_area_free(htab->buckets);
> @@ -2167,19 +2226,6 @@ static struct htab_elem *alloc_sock_hash_elem(struct bpf_htab *htab,
> return l_new;
> }
>
> -static struct htab_elem *lookup_elem_raw(struct hlist_head *head,
> - u32 hash, void *key, u32 key_size)
> -{
> - struct htab_elem *l;
> -
> - hlist_for_each_entry_rcu(l, head, hash_node) {
> - if (l->hash == hash && !memcmp(&l->key, key, key_size))
> - return l;
> - }
> -
> - return NULL;
> -}
> -
> static inline u32 htab_map_hash(const void *key, u32 key_len)
> {
> return jhash(key, key_len, 0);
> @@ -2307,8 +2353,10 @@ static int sock_hash_ctx_update_elem(struct bpf_sock_ops_kern *skops,
> psock = smap_psock_sk(l_old->sk);
>
> hlist_del_rcu(&l_old->hash_node);
> + write_lock_bh(&l_old->sk->sk_callback_lock);
> smap_list_remove(psock, NULL, l_old);
> smap_release_sock(psock, l_old->sk);
> + write_unlock_bh(&l_old->sk->sk_callback_lock);
> free_htab_elem(htab, l_old);
> }
> raw_spin_unlock_bh(&b->lock);
>
^ permalink raw reply
* [PATCH net-next v3 2/2] r8169: Reinstate ASPM Support
From: Kai-Heng Feng @ 2018-06-15 5:32 UTC (permalink / raw)
To: davem
Cc: ryankao, hayeswang, hau, hkallweit1, romieu, bhelgaas, acelan.kao,
netdev, linux-pci, linux-kernel, Kai-Heng Feng
In-Reply-To: <20180615053219.14053-1-kai.heng.feng@canonical.com>
On Intel Coffe Lake platforms, ASPM support in r8169 is the last missing
puzzle to let CPU's Package C-State reaches PC8.
Without ASPM support, the CPU cannot reach beyond PC3. PC8 can save
additional ~3W in comparison with PC3 on my testing platform, Dell G3
3779.
This is based on the work from Chunhao Lin <hau@realtek.com>.
Signed-off-by: Kai-Heng Feng <kai.heng.feng@canonical.com>
---
v3:
- Change commit message wording.
- Rename the function to rtl_hw_aspm_clkreq_enable().
v2:
- Remove module parameter.
- Remove pci_disable_link_state().
drivers/net/ethernet/realtek/r8169.c | 40 +++++++++++++++++++---------
1 file changed, 27 insertions(+), 13 deletions(-)
diff --git a/drivers/net/ethernet/realtek/r8169.c b/drivers/net/ethernet/realtek/r8169.c
index 9b55ce513a36..269ac7561368 100644
--- a/drivers/net/ethernet/realtek/r8169.c
+++ b/drivers/net/ethernet/realtek/r8169.c
@@ -5289,6 +5289,17 @@ static void rtl_pcie_state_l2l3_enable(struct rtl8169_private *tp, bool enable)
RTL_W8(tp, Config3, data);
}
+static void rtl_hw_aspm_clkreq_enable(struct rtl8169_private *tp, bool enable)
+{
+ if (enable) {
+ RTL_W8(tp, Config2, RTL_R8(tp, Config2) | ClkReqEn);
+ RTL_W8(tp, Config5, RTL_R8(tp, Config5) | ASPM_en);
+ } else {
+ RTL_W8(tp, Config2, RTL_R8(tp, Config2) & ~ClkReqEn);
+ RTL_W8(tp, Config5, RTL_R8(tp, Config5) & ~ASPM_en);
+ }
+}
+
static void rtl_hw_start_8168bb(struct rtl8169_private *tp)
{
RTL_W8(tp, Config3, RTL_R8(tp, Config3) & ~Beacon_en);
@@ -5645,9 +5656,9 @@ static void rtl_hw_start_8168g_1(struct rtl8169_private *tp)
rtl_hw_start_8168g(tp);
/* disable aspm and clock request before access ephy */
- RTL_W8(tp, Config2, RTL_R8(tp, Config2) & ~ClkReqEn);
- RTL_W8(tp, Config5, RTL_R8(tp, Config5) & ~ASPM_en);
+ rtl_hw_aspm_clkreq_enable(tp, false);
rtl_ephy_init(tp, e_info_8168g_1, ARRAY_SIZE(e_info_8168g_1));
+ rtl_hw_aspm_clkreq_enable(tp, true);
}
static void rtl_hw_start_8168g_2(struct rtl8169_private *tp)
@@ -5680,9 +5691,9 @@ static void rtl_hw_start_8411_2(struct rtl8169_private *tp)
rtl_hw_start_8168g(tp);
/* disable aspm and clock request before access ephy */
- RTL_W8(tp, Config2, RTL_R8(tp, Config2) & ~ClkReqEn);
- RTL_W8(tp, Config5, RTL_R8(tp, Config5) & ~ASPM_en);
+ rtl_hw_aspm_clkreq_enable(tp, false);
rtl_ephy_init(tp, e_info_8411_2, ARRAY_SIZE(e_info_8411_2));
+ rtl_hw_aspm_clkreq_enable(tp, true);
}
static void rtl_hw_start_8168h_1(struct rtl8169_private *tp)
@@ -5699,8 +5710,7 @@ static void rtl_hw_start_8168h_1(struct rtl8169_private *tp)
};
/* disable aspm and clock request before access ephy */
- RTL_W8(tp, Config2, RTL_R8(tp, Config2) & ~ClkReqEn);
- RTL_W8(tp, Config5, RTL_R8(tp, Config5) & ~ASPM_en);
+ rtl_hw_aspm_clkreq_enable(tp, false);
rtl_ephy_init(tp, e_info_8168h_1, ARRAY_SIZE(e_info_8168h_1));
RTL_W32(tp, TxConfig, RTL_R32(tp, TxConfig) | TXCFG_AUTO_FIFO);
@@ -5779,6 +5789,8 @@ static void rtl_hw_start_8168h_1(struct rtl8169_private *tp)
r8168_mac_ocp_write(tp, 0xe63e, 0x0000);
r8168_mac_ocp_write(tp, 0xc094, 0x0000);
r8168_mac_ocp_write(tp, 0xc09e, 0x0000);
+
+ rtl_hw_aspm_clkreq_enable(tp, true);
}
static void rtl_hw_start_8168ep(struct rtl8169_private *tp)
@@ -5830,11 +5842,12 @@ static void rtl_hw_start_8168ep_1(struct rtl8169_private *tp)
};
/* disable aspm and clock request before access ephy */
- RTL_W8(tp, Config2, RTL_R8(tp, Config2) & ~ClkReqEn);
- RTL_W8(tp, Config5, RTL_R8(tp, Config5) & ~ASPM_en);
+ rtl_hw_aspm_clkreq_enable(tp, false);
rtl_ephy_init(tp, e_info_8168ep_1, ARRAY_SIZE(e_info_8168ep_1));
rtl_hw_start_8168ep(tp);
+
+ rtl_hw_aspm_clkreq_enable(tp, true);
}
static void rtl_hw_start_8168ep_2(struct rtl8169_private *tp)
@@ -5846,14 +5859,15 @@ static void rtl_hw_start_8168ep_2(struct rtl8169_private *tp)
};
/* disable aspm and clock request before access ephy */
- RTL_W8(tp, Config2, RTL_R8(tp, Config2) & ~ClkReqEn);
- RTL_W8(tp, Config5, RTL_R8(tp, Config5) & ~ASPM_en);
+ rtl_hw_aspm_clkreq_enable(tp, false);
rtl_ephy_init(tp, e_info_8168ep_2, ARRAY_SIZE(e_info_8168ep_2));
rtl_hw_start_8168ep(tp);
RTL_W8(tp, DLLPR, RTL_R8(tp, DLLPR) & ~PFM_EN);
RTL_W8(tp, MISC_1, RTL_R8(tp, MISC_1) & ~PFM_D3COLD_EN);
+
+ rtl_hw_aspm_clkreq_enable(tp, true);
}
static void rtl_hw_start_8168ep_3(struct rtl8169_private *tp)
@@ -5867,8 +5881,7 @@ static void rtl_hw_start_8168ep_3(struct rtl8169_private *tp)
};
/* disable aspm and clock request before access ephy */
- RTL_W8(tp, Config2, RTL_R8(tp, Config2) & ~ClkReqEn);
- RTL_W8(tp, Config5, RTL_R8(tp, Config5) & ~ASPM_en);
+ rtl_hw_aspm_clkreq_enable(tp, false);
rtl_ephy_init(tp, e_info_8168ep_3, ARRAY_SIZE(e_info_8168ep_3));
rtl_hw_start_8168ep(tp);
@@ -5888,6 +5901,8 @@ static void rtl_hw_start_8168ep_3(struct rtl8169_private *tp)
data = r8168_mac_ocp_read(tp, 0xe860);
data |= 0x0080;
r8168_mac_ocp_write(tp, 0xe860, data);
+
+ rtl_hw_aspm_clkreq_enable(tp, true);
}
static void rtl_hw_start_8168(struct rtl8169_private *tp)
@@ -7646,7 +7661,6 @@ static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
mii->reg_num_mask = 0x1f;
mii->supports_gmii = cfg->has_gmii;
-
/* enable device (incl. PCI PM wakeup and hotplug setup) */
rc = pcim_enable_device(pdev);
if (rc < 0) {
--
2.17.0
^ permalink raw reply related
* [PATCH net-next v3 1/2] r8169: Don't disable ASPM in the driver
From: Kai-Heng Feng @ 2018-06-15 5:32 UTC (permalink / raw)
To: davem
Cc: ryankao, hayeswang, hau, hkallweit1, romieu, bhelgaas, acelan.kao,
netdev, linux-pci, linux-kernel, Kai-Heng Feng
Enable or disable ASPM should be done in PCI core instead of in the
device driver.
Commit ba04c7c93bbc ("r8169: disable ASPM") uses
pci_disable_link_state() to disable ASPM, but it's not the best way to
do it. If the device really wants to disable ASPM, we can use a quirk in
PCI core to prevent the PCI core from setting ASPM before probe.
Let's remove pci_disable_link_state() for now. Use PCI core quirks if
any regression happens.
Signed-off-by: Kai-Heng Feng <kai.heng.feng@canonical.com>
---
v3:
- Change commit message wording.
- Rename the function to rtl_hw_aspm_clkreq_enable().
v2:
- Remove module parameter.
- Remove pci_disable_link_state().
drivers/net/ethernet/realtek/r8169.c | 5 -----
1 file changed, 5 deletions(-)
diff --git a/drivers/net/ethernet/realtek/r8169.c b/drivers/net/ethernet/realtek/r8169.c
index 75dfac0248f4..9b55ce513a36 100644
--- a/drivers/net/ethernet/realtek/r8169.c
+++ b/drivers/net/ethernet/realtek/r8169.c
@@ -25,7 +25,6 @@
#include <linux/dma-mapping.h>
#include <linux/pm_runtime.h>
#include <linux/firmware.h>
-#include <linux/pci-aspm.h>
#include <linux/prefetch.h>
#include <linux/ipv6.h>
#include <net/ip6_checksum.h>
@@ -7647,10 +7646,6 @@ static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
mii->reg_num_mask = 0x1f;
mii->supports_gmii = cfg->has_gmii;
- /* disable ASPM completely as that cause random device stop working
- * problems as well as full system hangs for some PCIe devices users */
- pci_disable_link_state(pdev, PCIE_LINK_STATE_L0S | PCIE_LINK_STATE_L1 |
- PCIE_LINK_STATE_CLKPM);
/* enable device (incl. PCI PM wakeup and hotplug setup) */
rc = pcim_enable_device(pdev);
--
2.17.0
^ permalink raw reply related
* Re: [PATCH RFC v2] rhashtable: implement rhashtable_walk_peek() using rhashtable_walk_last_seen()
From: NeilBrown @ 2018-06-15 5:31 UTC (permalink / raw)
To: Herbert Xu, Tom Herbert
Cc: Thomas Graf, Linux Kernel Network Developers, LKML, Tom Herbert
In-Reply-To: <20180615042331.3dmqi3bumzk6lu5c@gondor.apana.org.au>
[-- Attachment #1: Type: text/plain, Size: 857 bytes --]
On Fri, Jun 15 2018, Herbert Xu wrote:
> On Thu, Jun 14, 2018 at 10:41:05AM -0700, Tom Herbert wrote:
>> On Mon, Jun 11, 2018 at 7:48 PM, NeilBrown <neilb@suse.com> wrote:
>> >
>> > rhashtable_walk_last_seen() does most of the work that
>> > rhashtable_walk_peek() needs done, so use it and put
>> > it in a "static inline".
>> > Also update the documentation for rhashtable_walk_peek() to clarify
>> > the expected use case.
>> >
>> > Signed-off-by: NeilBrown <neilb@suse.com>
>>
>> Acked-by: Tom Herbert <tom@quantonium.net>
>
> Acked-by: Herbert Xu <herbert@gondor.apana.org.au>
Thanks to your both.
I assume this is an implicit Ack for
rhashtable: add rhashtable_walk_last_seen()
I'll send out a patchset of all the acked patches shortly after the
merge window closes, and then see where we are up to.
Thanks,
NeilBrown
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 832 bytes --]
^ permalink raw reply
* Re: [PATCH bpf-net] selftests/bpf: delete xfrm tunnel when test exits.
From: Eyal Birger @ 2018-06-15 5:24 UTC (permalink / raw)
To: William Tu; +Cc: netdev, anders.roxell
In-Reply-To: <1528977666-26477-1-git-send-email-u9012063@gmail.com>
> On 14 Jun 2018, at 15:01, William Tu <u9012063@gmail.com> wrote:
>
> Make the printting of bpf xfrm tunnel better and
> cleanup xfrm state and policy when xfrm test finishes.
Yeah the ‘tee’ was useful when developing the test - I could see what’s going on :)
Now that it’s in ‘selftests’ it’s definitely better without it.
Thanks for the cleanup!
Eyal.
^ permalink raw reply
* Re: [PATCH net-next,RFC 00/13] New fast forwarding path
From: Steffen Klassert @ 2018-06-15 5:23 UTC (permalink / raw)
To: Willem de Bruijn; +Cc: Pablo Neira Ayuso, netfilter-devel, Network Development
In-Reply-To: <CAF=yD-+fJ_isaGtfLFVgMQ4_7K-v7b+Kyo6iOB-5kUbbPJXT7Q@mail.gmail.com>
On Thu, Jun 14, 2018 at 11:50:49AM -0400, Willem de Bruijn wrote:
> > This patchset supports both layer 3 IPv4 and IPv6, and layer 4 TCP and
> > UDP protocols. This fastpath also integrates with the IPSec
> > infrastructure and the ESP protocol.
> >
> > We have collected performance numbers:
> >
> > TCP TSO TCP Fast Forward
> > 32.5 Gbps 35.6 Gbps
> >
> > UDP UDP Fast Forward
> > 17.6 Gbps 35.6 Gbps
> >
> > ESP ESP Fast Forward
> > 6 Gbps 7.5 Gbps
> >
> > For UDP, this is doubling performance, and we almost achieve line rate
> > with one single CPU using the Intel i40e NIC. We got similar numbers
> > with the Mellanox ConnectX-4. For TCP, this is slightly improving things
> > even if TSO is being defeated given that we need to segment the packet
> > chain in software.
>
> The difference between TCP and UDP stems from lack of GRO for UDP.
Right.
> We
> recently added UDP GSO to allow for batch traversal of the UDP stack on
> transmission. Adding a UDP GRO handler can probably extend batching to
> the forwarding path in a similar way without the need for a new infrastructure.
That's more or less what we did. The batching method ist just
optimized for the forwarding path. We are generating skb chains
by chaning at the frag_list pointer of the first skb. With that,
we don't need to mange packet. We keep the packets in the native
form, so the 'segmentation' is rather easy.
The rest is just to be able to configure this and to make
sure that we handle only flows that are going to be (fast)
forwarded, as the upper stack can not (yet) handle such
skb chains.
^ permalink raw reply
* Re: [PATCH bpf-next v5 00/10] BTF: BPF Type Format
From: Yonghong Song @ 2018-06-15 4:56 UTC (permalink / raw)
To: Arnaldo Carvalho de Melo, Alexei Starovoitov
Cc: Martin KaFai Lau, netdev, Daniel Borkmann, kernel-team, Wang Nan,
Jiri Olsa, Namhyung Kim, Ingo Molnar
In-Reply-To: <20180614180017.GJ30043@kernel.org>
On 6/14/18 11:00 AM, Arnaldo Carvalho de Melo wrote:
> Em Thu, Jun 14, 2018 at 02:47:59PM -0300, Arnaldo Carvalho de Melo escreveu:
>> Em Thu, Jun 14, 2018 at 10:21:30AM -0700, Alexei Starovoitov escreveu:
>>> On 6/14/18 10:18 AM, Arnaldo Carvalho de Melo wrote:
>>>> Just out of curiosity, is there any plan to have this as a clang option?
>
>>> I think
>>> clang ... -mllvm -mattr=dwarfris
>>> should work.
>
>> The message "(LLVM option parsing)" implies what you suggest, but didn't
>> worked :-\
>
>> -mllvm <value> Additional arguments to forward to LLVM's option processing
>
>> Almost there tho :-\
>
> So I thought that this -mattr=dwarfris would be available only after I
> set the target, because I tried 'llc -mattr=help' and dwarfris wasn't
> there:
>
> [acme@jouet perf]$ llc -mattr=help |& grep dwarf
> [acme@jouet perf]$
>
> Only after I set the arch it appears:
>
> [acme@jouet perf]$ llc -march=bpf -mattr=help |& grep dwarf
> dwarfris - Disable MCAsmInfo DwarfUsesRelocationsAcrossSections.
> dwarfris - Disable MCAsmInfo DwarfUsesRelocationsAcrossSections.
> dwarfris - Disable MCAsmInfo DwarfUsesRelocationsAcrossSections.
> [acme@jouet perf]$
>
> But even after moving the '-mllvm -mattr=dwarfris' to after '-target
> bpf' it still can't grok it :-\
>
> /usr/local/bin/clang -D__KERNEL__ -D__NR_CPUS__=4 -DLINUX_VERSION_CODE=0x41100 -g -nostdinc -isystem /usr/lib/gcc/x86_64-redhat-linux/7/include -I/home/acme/git/linux/arch/x86/include -I./arch/x86/include/generated -I/home/acme/git/linux/include -I./include -I/home/acme/git/linux/arch/x86/include/uapi -I./arch/x86/include/generated/uapi -I/home/acme/git/linux/include/uapi -I./include/generated/uapi -include /home/acme/git/linux/include/linux/kconfig.h -I/home/acme/lib/include/perf/bpf -Wno-unused-value -Wno-pointer-sign -working-directory /lib/modules/4.17.0-rc5/build -c /home/acme/bpf/hello.c -target bpf -mllvm -mattr=dwarfris -O2 -o hello.o
>
> So onlye with 'clang ... -target bpf -emit-llvm -O2 -o - | llc -march=bpf -mattr=dwarfris ...'
> things work as we expect.
Right. Currently, the only way to use option -mattr=dwarfris is through
llc. The "clang -mllvm -mattr=dwarfris" won't work since
-mllvm <value> Additional arguments to forward to LLVM's option
processing
and -mattr=dwarfris is not in LLVM auto option processing system.
Those options, in llvm source code, typically have a pattern like below:
===
static cl::opt<unsigned> MemCmpEqZeroNumLoadsPerBlock(
"memcmp-num-loads-per-block", cl::Hidden, cl::init(1),
cl::desc("The number of loads per basic block for inline expansion of "
"memcmp that is only being compared against zero."));
===
I really want to get rid of this option as well. To make pahole work
with the default default format, I need to add bpf support to
libdwfl in elfutils repo. I will work on that.
> - Arnaldo
>
^ 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