* Re: [PATCH v2 1/7] Basic kernel memory functionality for the Memory Controller
From: Glauber Costa @ 2011-09-18 3:39 UTC (permalink / raw)
To: Kirill A. Shutemov
Cc: linux-kernel, paul, lizf, kamezawa.hiroyu, ebiederm, davem,
gthelen, netdev, linux-mm
In-Reply-To: <20110917174535.GA1658@shutemov.name>
>> struct mem_cgroup_stat_cpu *stat;
>> @@ -391,6 +404,7 @@ enum charge_type {
>> #define _MEM (0)
>> #define _MEMSWAP (1)
>> #define _OOM_TYPE (2)
>> +#define _KMEM (3)
>
> Ditto. Can we use enum instead?
Yes we can (tm)
>> if (!mem_cgroup_is_root(mem)) {
>> if (!swap)
>> - return res_counter_read_u64(&mem->res, RES_USAGE);
>> + kmem += res_counter_read_u64(&mem->res, RES_USAGE);
>> else
>> - return res_counter_read_u64(&mem->memsw, RES_USAGE);
>> + kmem += res_counter_read_u64(&mem->memsw, RES_USAGE);
>> +
>> + return kmem;
>> }
>>
>> val = mem_cgroup_recursive_stat(mem, MEM_CGROUP_STAT_CACHE);
>
> No kernel memory accounting for root cgroup, right?
Not sure. Maybe kernel memory accounting is useful even for root cgroup.
Same as normal memory accounting... what we want to avoid is kernel
memory limits. OTOH, if we are not limiting it anyway, accounting it is
just useless overhead... Even the statistics can then be gathered
through all
the proc files that show slab usage, I guess?
>
>> @@ -3979,6 +3999,10 @@ static u64 mem_cgroup_read(struct cgroup *cont, struct cftype *cft)
>> else
>> val = res_counter_read_u64(&mem->memsw, name);
>> break;
>> + case _KMEM:
>> + val = res_counter_read_u64(&mem->kmem, name);
>> + break;
>> +
>
> Always zero in root cgroup?
Yes, if we're not accounting, it should be zero. WARN_ON, maybe?
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply
* [GIT] Networking
From: David Miller @ 2011-09-18 6:21 UTC (permalink / raw)
To: torvalds; +Cc: akpm, netdev, linux-kernel
I thought the kernel.org issue would be cleared up a lot sooner than
it appears it will end up being resolved, so I held off on creating
another GIT tree to push changes to you until just a day or two ago.
Anyways, here are all of the fixes which have queued up meanwhile.
As usual, the majority of the stuff is in the drivers, with only a few
protocol and generic networking fixes here and there.
1) Fix DMA mapping failures and checksum offload failure handling in
ibmveth driver, from Anton Blanchard and Brian King.
2) Netfilter fixes:
a) Fix xt_rateest_mt_checkentry() error handling, from Eric Dumazet.
b) Fix leak in ipq_build_packet_message(), from Jesper Juhl.
c) Build dependencies for ebtables are wrong, from Bart De Schuymer.
d) Userspace should never be allowed to send NF_STOLEN verdicts to
nf_queue, fix from Florian Westphal.
e) Fix translation of peer's call-id field in nf_ct_pptp when
both the server and client are NATed, from Sanket Shah.
f) Fix handling of invalid TCP options in nf_ct_tcp, from Jozsef Kadlecsik.
g) Fix TCP option lengh calculations in nf_ct_tcp's tcp_sack(), also
from Jozsef.
3) Socket address hardening in inet_bind() broke some apps, loosen
the checks so they work again.
4) Fix locking in mac80211's __sta_info_destroy() from Johannes Berg.
5) Fix crash when unplugging ath9k devices, from Mohammed Shafi Shajakhan.
6) Fix stack corruption and command queue timeouts in iwlagn, from Johannes
Berg.
7) Fix resets while reading device config, eliminate accidental accesses of
not-present registers, and fix coalescing bug in bnx2x driver from Dmitry Kravkov,
Vladislav Zolotarov, and Ariel Elior.
8) cls_rsvp.h indexes attribute array off-by-one, fix from Igor Maravić.
9) Syn flooding messages fill people's logs, rate limit it and report the
number of times it is detected in a statistic. From Eric Dumazet.
10) CAN protocol layer needs to use del_time_sync(), from Rajan Aggarwal.
11) Flow cache accidently uses entries from one namespace in another, fix
from David Ward.
12) If zero copy skbs hit forward path, we must copy them, from Michael S. Tsirkin.
13) ether_setup() needs to "OR" in the IFF_TX_SKB_SHARING bit so that other
settings made already don't get lost, from Neil Horman.
14) Revert regression causing SFC write-combining change, from Ben Hutchings.
15) Cure FIFO overrun handling in pch_gbe driver, from Toshiharu Okada.
16) SCTP connection wedges when it sees two COOKIE_ECHO chunks, fix from
Max Matveev.
17) Fix NULL deref in CAIF stack, from Eric Dumazet.
18) Fix ipv4 route metrics leak caused by kfree_rcu() conversions, fix
from Zheng Yan.
19) flow_key_compare has dependencies upon flow key alignment, make sure
they are met. From David Ward.
20) Also from David Ward, compare flow keys correctly based upon their
address-family specific size. Otherwise we reference garbage.
21) Fix ti_hecc driver build, missing include, from Daniel Mack.
22) Don't use inetpeer for the metrics of anything other than host
routes, from Zheng Yan.
23) Fix EEPROM signature handling of e1000 devices on HPPA. From
Jeff Kirsher.
24) ipv6's datagram_send_ctl() needs to adhere to the socket's
->transparent setting, fix from Maciej Żenczykowski.
Please pull, thanks a lot!
The following changes since commit 9e79e3e9dd9672b37ac9412e9a926714306551fe:
Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc (2011-08-30 11:28:18 -0700)
are available in the git repository at:
git://github.com/davem330/net.git master
Andi Kleen (1):
IRDA: Fix global type conflicts in net/irda/irsysctl.c v2
Anton Blanchard (3):
ibmveth: Fix issue with DMA mapping failure
ibmveth: Checksum offload is always disabled
ibmveth: Fix checksum offload failure handling
Ariel Elior (1):
bnx2x: Fix for a host coalescing bug which impared latency.
Bart De Schuymer (1):
netfilter: ebtables: fix ebtables build dependency
Ben Hutchings (2):
Revert "sfc: Use write-combining to reduce TX latency" and follow-ups
sfc: Use 64-bit writes for TX push where possible
Brian King (1):
ibmveth: Fix DMA unmap error
Daniel Mack (1):
can: ti_hecc: include linux/io.h
David S. Miller (3):
Merge branch 'master' of git://git.kernel.org/.../kaber/nf-2.6
Merge branch 'for-davem' of git://git.infradead.org/users/linville/wireless
Merge branch 'master' of ../netdev/
David Ward (1):
net: Align AF-specific flowi structs to long
Dmitry Kravkov (6):
bnx2x: fix BRB thresholds for dropless_fc mode
bnx2x: decrease print level to debug
bnx2x: fix MF for 4-port devices
bnx2x: don't reset device while reading its configuration.
bnx2x: init fw_seq after undi_unload is done
bnx2x: don't access removed registers on 57712 and above
Eric Dumazet (4):
netfilter: xt_rateest: fix xt_rateest_mt_checkentry()
net: ipv4: relax AF_INET check in bind()
tcp: Change possible SYN flooding messages
caif: fix a potential NULL dereference
Felix Fietkau (1):
ath9k_hw: fix calibration on 5 ghz
Florian Westphal (1):
netfilter: nf_queue: reject NF_STOLEN verdicts from userspace
George (2):
rtlwifi: rtl8192su: Fix problem connecting to HT-enabled AP
rtlwifi: Fix problem when switching connections
Ian Campbell (3):
MAINTAINERS: Update ATLX driver maintainers
MAINTAINERS: Update Cisco VIC driver maintainers
MAINTAINERS: Update BNA 10G Maintainer
Igor Maravić (1):
pkt_sched: cls_rsvp.h was outdated
Jeff Kirsher (1):
e1000: Fix driver to be used on PA RISC C8000 workstations
Jesper Juhl (1):
netfilter: ip_queue: Fix small leak in ipq_build_packet_message()
Johannes Berg (3):
mac80211: fix missing sta_lock in __sta_info_destroy
iwlagn: fix stack corruption
iwlagn: fix command queue timeout
John W. Linville (1):
Merge branch 'master' of git://git.infradead.org/users/linville/wireless into for-davem
Jozsef Kadlecsik (2):
netfilter: nf_ct_tcp: fix incorrect handling of invalid TCP option
netfilter: nf_ct_tcp: wrong multiplication of TCPOLEN_TSTAMP_ALIGNED in tcp_sack skips fastpath
Kavan Smith (1):
ipheth: iPhone 4 Verizon CDMA USB Product ID add
Maciej Żenczykowski (1):
net: relax PKTINFO non local ipv6 udp xmit check
Marcos Paulo de Souza (1):
Documentation: networking: dmfe.txt: Remove the maintainer of orphan networking driver
Max Matveev (1):
sctp: deal with multiple COOKIE_ECHO chunks
Michael S. Tsirkin (1):
net: copy userspace buffers on device forwarding
Mohammed Shafi Shajakhan (1):
ath9k: Fix kernel panic on unplugging the device
Patrick McHardy (1):
netfilter: update netfilter git URL
Rajkumar Manoharan (1):
ath9k_hw: Fix init mode register regression
Sanket Shah (1):
netfilter: nf_ct_pptp: fix DNATed PPTP connection address translation
Stanislaw Gruszka (1):
iwlegacy: fix BUG_ON(info->control.rates[0].idx < 0)
Toshiharu Okada (3):
pch_gbe: fixed the issue which receives an unnecessary packet.
pch_gbe: added the process of FIFO over run error
pch_gbe: support ML7831 IOH
Vladislav Zolotarov (1):
bnx2x: fix rx ring size report
Yan, Zheng (2):
ipv4: Fix fib_info->fib_metrics leak
ipv6: don't use inetpeer to store metrics for routes.
Yaniv Rosner (6):
bnx2x: Fix ETS bandwidth
bnx2x: Enable FEC for 57810-KR
bnx2x: Remove fiber remote fault detection
bnx2x: Fix XMAC loopback test
bnx2x: Fix 578xx link LED
bnx2x: Fix ethtool advertisement
dpward (2):
net: Make flow cache namespace-aware
net: Handle different key sizes between address families in flow cache
nhorman (1):
net: don't clear IFF_XMIT_DST_RELEASE in ether_setup
rajan.aggarwal85@gmail.com (1):
net/can/af_can.c: Change del_timer to del_timer_sync
Documentation/networking/dmfe.txt | 3 +-
MAINTAINERS | 6 +-
drivers/net/Kconfig | 11 +-
drivers/net/bnx2x/bnx2x.h | 124 +++++++---
drivers/net/bnx2x/bnx2x_cmn.c | 27 +-
drivers/net/bnx2x/bnx2x_ethtool.c | 48 +++-
drivers/net/bnx2x/bnx2x_link.c | 46 ++--
drivers/net/bnx2x/bnx2x_main.c | 162 ++++++++----
drivers/net/bnx2x/bnx2x_reg.h | 7 +-
drivers/net/bnx2x/bnx2x_stats.c | 7 +-
drivers/net/can/ti_hecc.c | 1 +
drivers/net/e1000/e1000_hw.c | 6 +
drivers/net/ibmveth.c | 48 +++--
drivers/net/pch_gbe/pch_gbe.h | 12 +-
drivers/net/pch_gbe/pch_gbe_main.c | 302 ++++++++++++++--------
drivers/net/sfc/efx.c | 18 +--
drivers/net/sfc/io.h | 6 -
drivers/net/sfc/mcdi.c | 46 ++--
drivers/net/sfc/nic.c | 7 -
drivers/net/sfc/nic.h | 2 -
drivers/net/sfc/siena.c | 25 +--
drivers/net/sfc/workarounds.h | 2 -
drivers/net/usb/ipheth.c | 5 +
drivers/net/wireless/ath/ath9k/ar9002_calib.c | 3 +-
drivers/net/wireless/ath/ath9k/ar9003_phy.c | 2 +-
drivers/net/wireless/ath/ath9k/main.c | 6 +
drivers/net/wireless/iwlegacy/iwl-3945-rs.c | 13 +-
drivers/net/wireless/iwlwifi/iwl-agn-ucode.c | 2 +-
drivers/net/wireless/iwlwifi/iwl-trans-tx-pcie.c | 2 +
drivers/net/wireless/rtlwifi/core.c | 8 +
drivers/net/wireless/rtlwifi/rtl8192cu/trx.c | 11 +-
include/linux/skbuff.h | 1 +
include/linux/snmp.h | 2 +
include/net/flow.h | 25 ++-
include/net/request_sock.h | 3 +-
include/net/sctp/command.h | 1 +
include/net/tcp.h | 3 +
include/net/transp_v6.h | 1 +
net/bridge/netfilter/Kconfig | 2 +-
net/caif/caif_dev.c | 6 +-
net/can/af_can.c | 2 +-
net/core/dev.c | 8 +
net/core/flow.c | 36 ++-
net/core/skbuff.c | 22 ++-
net/ethernet/eth.c | 2 +-
net/ipv4/af_inet.c | 7 +-
net/ipv4/fib_semantics.c | 10 +-
net/ipv4/netfilter/ip_queue.c | 12 +-
net/ipv4/proc.c | 2 +
net/ipv4/tcp_ipv4.c | 51 ++--
net/ipv6/datagram.c | 5 +-
net/ipv6/ip6_flowlabel.c | 8 +-
net/ipv6/ipv6_sockglue.c | 2 +-
net/ipv6/netfilter/ip6_queue.c | 12 +-
net/ipv6/raw.c | 4 +-
net/ipv6/route.c | 33 ++-
net/ipv6/tcp_ipv6.c | 31 +--
net/ipv6/udp.c | 4 +-
net/irda/irsysctl.c | 6 +-
net/irda/qos.c | 6 +-
net/mac80211/sta_info.c | 2 +-
net/netfilter/nf_conntrack_pptp.c | 1 +
net/netfilter/nf_conntrack_proto_tcp.c | 6 +-
net/netfilter/nfnetlink_queue.c | 4 +-
net/netfilter/xt_rateest.c | 9 +-
net/sched/cls_rsvp.h | 27 +-
net/sctp/sm_sideeffect.c | 5 +
net/sctp/sm_statefuns.c | 6 +
68 files changed, 827 insertions(+), 508 deletions(-)
^ permalink raw reply
* Re: [PATCH ] dynamic_debug: call __netdev_printk only for CONFIG_NET
From: Greg KH @ 2011-09-18 8:27 UTC (permalink / raw)
To: Jason Baron
Cc: Arnd Bergmann, gregkh, joe, jim.cromie, bvanassche, linux-kernel,
davem, aloisio.almeida, netdev
In-Reply-To: <20110901151817.GA14324@redhat.com>
On Thu, Sep 01, 2011 at 11:18:18AM -0400, Jason Baron wrote:
> On Thu, Sep 01, 2011 at 04:57:02PM +0200, Arnd Bergmann wrote:
> > __netdev_printk is only defined when CONFIG_NET is set. Since we only need
> > __dynamic_netdev_dbg for network drivers, we can make it conditional on the
> > same Kconfig symbol.
> >
>
> Hi,
>
> Yes, I've posted a fix for this:
>
> https://lkml.org/lkml/2011/8/30/297
>
> Hopefully, it will be pulled in soon.
As that thread again spun off into confusion, can you please resend the
end result?
thanks,
greg k-h
^ permalink raw reply
* [PATCH 0/4] add fec support for imx6q
From: Shawn Guo @ 2011-09-18 11:54 UTC (permalink / raw)
To: netdev; +Cc: David S. Miller, linux-arm-kernel, patches
This series adds imx6q enet support. The imx6q enet is a derivative of
imx28 enet controller. It fixed the frame endian issue found on imx28,
and added 1 Gbps support.
It's based on v3.1-rc6.
Shawn Guo (4):
net/fec: change phy-reset-gpio request warning to debug message
net/fec: fix fec1 check in fec_enet_mii_init()
net/fec: set phy_speed to the optimal frequency 2.5 MHz
net/fec: add imx6q enet support
drivers/net/Kconfig | 3 +-
drivers/net/fec.c | 65 ++++++++++++++++++++++++++++++++++++++++-----------
2 files changed, 52 insertions(+), 16 deletions(-)
^ permalink raw reply
* [PATCH 1/4] net/fec: change phy-reset-gpio request warning to debug message
From: Shawn Guo @ 2011-09-18 11:54 UTC (permalink / raw)
To: netdev; +Cc: David S. Miller, linux-arm-kernel, patches, Shawn Guo
In-Reply-To: <1316346852-17090-1-git-send-email-shawn.guo@linaro.org>
FEC can work without a phy reset on some platforms, which means not
very platform necessarily have a phy-reset gpio encoded in device tree.
So it makes more sense to have the phy-reset-gpio request failure as
a debug message rather than a warning.
Signed-off-by: Shawn Guo <shawn.guo@linaro.org>
---
drivers/net/fec.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/drivers/net/fec.c b/drivers/net/fec.c
index e8266cc..6a638e9 100644
--- a/drivers/net/fec.c
+++ b/drivers/net/fec.c
@@ -1422,7 +1422,7 @@ static int __devinit fec_reset_phy(struct platform_device *pdev)
phy_reset = of_get_named_gpio(np, "phy-reset-gpios", 0);
err = gpio_request_one(phy_reset, GPIOF_OUT_INIT_LOW, "phy-reset");
if (err) {
- pr_warn("FEC: failed to get gpio phy-reset: %d\n", err);
+ pr_debug("FEC: failed to get gpio phy-reset: %d\n", err);
return err;
}
msleep(1);
--
1.7.4.1
^ permalink raw reply related
* [PATCH 2/4] net/fec: fix fec1 check in fec_enet_mii_init()
From: Shawn Guo @ 2011-09-18 11:54 UTC (permalink / raw)
To: netdev; +Cc: David S. Miller, linux-arm-kernel, patches, Shawn Guo
In-Reply-To: <1316346852-17090-1-git-send-email-shawn.guo@linaro.org>
In function fec_enet_mii_init(), it uses non-zero pdev->id as part
of the condition to check the second fec instance (fec1). This works
before the driver supports device tree probe. But in case of device
tree probe, pdev->id is -1 which is also non-zero, so the logic becomes
broken when device tree probe gets supported.
The patch change the logic to check "pdev->id > 0" as the part of the
condition for identifying fec1.
Signed-off-by: Shawn Guo <shawn.guo@linaro.org>
---
drivers/net/fec.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/drivers/net/fec.c b/drivers/net/fec.c
index 6a638e9..5ef0e34 100644
--- a/drivers/net/fec.c
+++ b/drivers/net/fec.c
@@ -996,7 +996,7 @@ static int fec_enet_mii_init(struct platform_device *pdev)
* mdio interface in board design, and need to be configured by
* fec0 mii_bus.
*/
- if ((id_entry->driver_data & FEC_QUIRK_ENET_MAC) && pdev->id) {
+ if ((id_entry->driver_data & FEC_QUIRK_ENET_MAC) && pdev->id > 0) {
/* fec1 uses fec0 mii_bus */
fep->mii_bus = fec0_mii_bus;
return 0;
--
1.7.4.1
^ permalink raw reply related
* [PATCH 3/4] net/fec: set phy_speed to the optimal frequency 2.5 MHz
From: Shawn Guo @ 2011-09-18 11:54 UTC (permalink / raw)
To: netdev; +Cc: David S. Miller, linux-arm-kernel, patches, Shawn Guo
In-Reply-To: <1316346852-17090-1-git-send-email-shawn.guo@linaro.org>
With the unnecessary 1 bit left-shift on fep->phy_speed during the
calculation, the phy_speed always runs at the half frequency of the
optimal one 2.5 MHz.
The patch removes that 1 bit left-shift to get the optimal phy_speed.
Signed-off-by: Shawn Guo <shawn.guo@linaro.org>
---
drivers/net/fec.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/drivers/net/fec.c b/drivers/net/fec.c
index 5ef0e34..04206e4 100644
--- a/drivers/net/fec.c
+++ b/drivers/net/fec.c
@@ -1007,7 +1007,7 @@ static int fec_enet_mii_init(struct platform_device *pdev)
/*
* Set MII speed to 2.5 MHz (= clk_get_rate() / 2 * phy_speed)
*/
- fep->phy_speed = DIV_ROUND_UP(clk_get_rate(fep->clk), 5000000) << 1;
+ fep->phy_speed = DIV_ROUND_UP(clk_get_rate(fep->clk), 5000000);
writel(fep->phy_speed, fep->hwp + FEC_MII_SPEED);
fep->mii_bus = mdiobus_alloc();
--
1.7.4.1
^ permalink raw reply related
* [PATCH 4/4] net/fec: add imx6q enet support
From: Shawn Guo @ 2011-09-18 11:54 UTC (permalink / raw)
To: netdev; +Cc: David S. Miller, linux-arm-kernel, patches, Shawn Guo
In-Reply-To: <1316346852-17090-1-git-send-email-shawn.guo@linaro.org>
The imx6q enet is a derivative of imx28 enet controller. It fixed
the frame endian issue found on imx28, and added 1 Gbps support.
Signed-off-by: Shawn Guo <shawn.guo@linaro.org>
---
drivers/net/Kconfig | 3 +-
drivers/net/fec.c | 59 +++++++++++++++++++++++++++++++++++++++++---------
2 files changed, 49 insertions(+), 13 deletions(-)
diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig
index 8d0314d..fb38962 100644
--- a/drivers/net/Kconfig
+++ b/drivers/net/Kconfig
@@ -1937,8 +1937,7 @@ config DECLANCE
config FEC
bool "FEC ethernet controller (of ColdFire and some i.MX CPUs)"
depends on M523x || M527x || M5272 || M528x || M520x || M532x || \
- IMX_HAVE_PLATFORM_FEC || MXS_HAVE_PLATFORM_FEC
- default IMX_HAVE_PLATFORM_FEC || MXS_HAVE_PLATFORM_FEC if ARM
+ ARCH_MXC || ARCH_MXS
select PHYLIB
help
Say Y here if you want to use the built-in 10/100 Fast ethernet
diff --git a/drivers/net/fec.c b/drivers/net/fec.c
index 04206e4..849cb0b 100644
--- a/drivers/net/fec.c
+++ b/drivers/net/fec.c
@@ -18,7 +18,7 @@
* Bug fixes and cleanup by Philippe De Muyter (phdm@macqel.be)
* Copyright (c) 2004-2006 Macq Electronique SA.
*
- * Copyright (C) 2010 Freescale Semiconductor, Inc.
+ * Copyright (C) 2010-2011 Freescale Semiconductor, Inc.
*/
#include <linux/module.h>
@@ -72,6 +72,10 @@
#define FEC_QUIRK_SWAP_FRAME (1 << 1)
/* Controller uses gasket */
#define FEC_QUIRK_USE_GASKET (1 << 2)
+/* Controller has GBIT support */
+#define FEC_QUIRK_HAS_GBIT (1 << 3)
+/* Controller's phy_speed bit field need to minus one */
+#define FEC_QUIRK_PHY_SPEED_MINUS_ONE (1 << 4)
static struct platform_device_id fec_devtype[] = {
{
@@ -88,6 +92,10 @@ static struct platform_device_id fec_devtype[] = {
.name = "imx28-fec",
.driver_data = FEC_QUIRK_ENET_MAC | FEC_QUIRK_SWAP_FRAME,
}, {
+ .name = "imx6q-fec",
+ .driver_data = FEC_QUIRK_ENET_MAC | FEC_QUIRK_HAS_GBIT |
+ FEC_QUIRK_PHY_SPEED_MINUS_ONE,
+ }, {
/* sentinel */
}
};
@@ -97,12 +105,14 @@ enum imx_fec_type {
IMX25_FEC = 1, /* runs on i.mx25/50/53 */
IMX27_FEC, /* runs on i.mx27/35/51 */
IMX28_FEC,
+ IMX6Q_FEC,
};
static const struct of_device_id fec_dt_ids[] = {
{ .compatible = "fsl,imx25-fec", .data = &fec_devtype[IMX25_FEC], },
{ .compatible = "fsl,imx27-fec", .data = &fec_devtype[IMX27_FEC], },
{ .compatible = "fsl,imx28-fec", .data = &fec_devtype[IMX28_FEC], },
+ { .compatible = "fsl,imx6q-fec", .data = &fec_devtype[IMX6Q_FEC], },
{ /* sentinel */ }
};
MODULE_DEVICE_TABLE(of, fec_dt_ids);
@@ -373,6 +383,7 @@ fec_restart(struct net_device *ndev, int duplex)
int i;
u32 temp_mac[2];
u32 rcntl = OPT_FRAME_SIZE | 0x04;
+ u32 ecntl = 0x2; /* ETHEREN */
/* Whack a reset. We should wait for this. */
writel(1, fep->hwp + FEC_ECNTRL);
@@ -442,18 +453,23 @@ fec_restart(struct net_device *ndev, int duplex)
/* Enable flow control and length check */
rcntl |= 0x40000000 | 0x00000020;
- /* MII or RMII */
+ /* RGMII, RMII or MII */
+ if (fep->phy_interface == PHY_INTERFACE_MODE_RGMII)
+ rcntl |= (1 << 6);
if (fep->phy_interface == PHY_INTERFACE_MODE_RMII)
rcntl |= (1 << 8);
else
rcntl &= ~(1 << 8);
- /* 10M or 100M */
- if (fep->phy_dev && fep->phy_dev->speed == SPEED_100)
- rcntl &= ~(1 << 9);
- else
- rcntl |= (1 << 9);
-
+ /* 1G, 100M or 10M */
+ if (fep->phy_dev) {
+ if (fep->phy_dev->speed == SPEED_1000)
+ ecntl |= (1 << 8);
+ else if (fep->phy_dev->speed == SPEED_100)
+ rcntl &= ~(1 << 9);
+ else
+ rcntl |= (1 << 9);
+ }
} else {
#ifdef FEC_MIIGSK_ENR
if (id_entry->driver_data & FEC_QUIRK_USE_GASKET) {
@@ -478,8 +494,15 @@ fec_restart(struct net_device *ndev, int duplex)
}
writel(rcntl, fep->hwp + FEC_R_CNTRL);
+ if (id_entry->driver_data & FEC_QUIRK_ENET_MAC) {
+ /* enable ENET endian swap */
+ ecntl |= (1 << 8);
+ /* enable ENET store and forward mode */
+ writel(1 << 8, fep->hwp + FEC_X_WMRK);
+ }
+
/* And last, enable the transmit and receive processing */
- writel(2, fep->hwp + FEC_ECNTRL);
+ writel(ecntl, fep->hwp + FEC_ECNTRL);
writel(0, fep->hwp + FEC_R_DES_ACTIVE);
/* Enable interrupts we wish to service */
@@ -490,6 +513,8 @@ static void
fec_stop(struct net_device *ndev)
{
struct fec_enet_private *fep = netdev_priv(ndev);
+ const struct platform_device_id *id_entry =
+ platform_get_device_id(fep->pdev);
/* We cannot expect a graceful transmit stop without link !!! */
if (fep->link) {
@@ -504,6 +529,10 @@ fec_stop(struct net_device *ndev)
udelay(10);
writel(fep->phy_speed, fep->hwp + FEC_MII_SPEED);
writel(FEC_DEFAULT_IMASK, fep->hwp + FEC_IMASK);
+
+ /* We have to keep ENET enabled to have MII interrupt stay working */
+ if (id_entry->driver_data & FEC_QUIRK_ENET_MAC)
+ writel(2, fep->hwp + FEC_ECNTRL);
}
@@ -918,6 +947,8 @@ static int fec_enet_mdio_reset(struct mii_bus *bus)
static int fec_enet_mii_probe(struct net_device *ndev)
{
struct fec_enet_private *fep = netdev_priv(ndev);
+ const struct platform_device_id *id_entry =
+ platform_get_device_id(fep->pdev);
struct phy_device *phy_dev = NULL;
char mdio_bus_id[MII_BUS_ID_SIZE];
char phy_name[MII_BUS_ID_SIZE + 3];
@@ -949,14 +980,18 @@ static int fec_enet_mii_probe(struct net_device *ndev)
snprintf(phy_name, MII_BUS_ID_SIZE, PHY_ID_FMT, mdio_bus_id, phy_id);
phy_dev = phy_connect(ndev, phy_name, &fec_enet_adjust_link, 0,
- PHY_INTERFACE_MODE_MII);
+ fep->phy_interface);
if (IS_ERR(phy_dev)) {
printk(KERN_ERR "%s: could not attach to PHY\n", ndev->name);
return PTR_ERR(phy_dev);
}
/* mask with MAC supported features */
- phy_dev->supported &= PHY_BASIC_FEATURES;
+ if (id_entry->driver_data & FEC_QUIRK_HAS_GBIT)
+ phy_dev->supported &= PHY_GBIT_FEATURES;
+ else
+ phy_dev->supported &= PHY_BASIC_FEATURES;
+
phy_dev->advertising = phy_dev->supported;
fep->phy_dev = phy_dev;
@@ -1008,6 +1043,8 @@ static int fec_enet_mii_init(struct platform_device *pdev)
* Set MII speed to 2.5 MHz (= clk_get_rate() / 2 * phy_speed)
*/
fep->phy_speed = DIV_ROUND_UP(clk_get_rate(fep->clk), 5000000);
+ if (id_entry->driver_data & FEC_QUIRK_PHY_SPEED_MINUS_ONE)
+ fep->phy_speed--;
writel(fep->phy_speed, fep->hwp + FEC_MII_SPEED);
fep->mii_bus = mdiobus_alloc();
--
1.7.4.1
^ permalink raw reply related
* (unknown)
From: Mrs. Tessy Brown @ 2011-09-18 13:54 UTC (permalink / raw)
Win $500,000 in Coca Cola Company @West Africa end of year
promo.To qualify, Email the correct answer to the question given below
to Mr. Frank Morris via(frankmorris10@ymail.com)
Question: Who won the 2010 FIFA World Cup in South Africa?
(A)England (B)Spain (C)Germany (D)Brazil.
^ permalink raw reply
* Re: [PATCH 1/2] iwlegacy: change IWL_WARN to IWL_DEBUG_HT in iwl4965_tx_agg_start
From: Greg Dietsche @ 2011-09-18 14:41 UTC (permalink / raw)
To: Stanislaw Gruszka; +Cc: linville, linux-wireless, netdev, linux-kernel
In-Reply-To: <20110908161129.GC2195@redhat.com>
On Thu, 8 Sep 2011 18:11:30 +0200, Stanislaw Gruszka wrote:
<snip>
> The second patch is ok. I'm not sure about first one, but we can get
> rid of "ctx = il_rxon_ctx_from_vif(vif)" at all, because we have
> only one context. Removing il_rxon_context structure from
> iwlegacy driver is my long term plan, you can look at that
> if you wish.
Sounds good. I'll let you know if I start to remove the il_rxon_context
structure. It looks right now like I might not have much free time in
the near future.
Greg
^ permalink raw reply
* Follow-up to routing IPv6 source address selection bug in kernel
From: Xavier Roche @ 2011-09-18 14:11 UTC (permalink / raw)
To: netdev
[ Moved from linux-kernel where it did not belong ]
Hi folks,
I reported a year ago a bug regarding source address selection in the
kernel for Ipv6, but it seem to be still there. If anyone has any
insightful advice on possible workarounds, or (better) possible fixes,
it would be great.
Basically, the "src" attribute of "ip -6 route add" is ignored, and
default source address selection is selected by the kernel.
This is probably related to the way the kernel handles RFC 3484 source
address selection [ The RFC states that [RFC 3484] "If the eight [source
address selection] rules fail to choose a single address, some
unspecified tie-breaker should be used". The unspecified tie-breaker
would then be the src routing information, or any additional netfilter
setting. ]
Selecting the source address according to outgoing parameters
(destination network, destination protocol, for example, but it could be
running uid/gid with advanced netfilter rules) is kind of handy when you
want to have dedicated addresses for, say, outgoing SMTP, outgoing HTTP,
outgoing SSH and so on..
This is especially true with IPv6: the default allocated size is at
least 16 billions billions IP addresses. Being able to use more than one
address per server is then kind of handy.
Binding to a special IP address for outgoing connections is difficult in
most cases, because the application would have to do the logic the
kernel is computing normally (destination on local network ? or on the
same interface ..) and would prevent proper use when multiple
interfaces/networks are in use.
The simplest way to achieve that would be to build a dedicated route for
a specific netblock, for example (this would not solve the
"per-destination-protocol" case, but this is a beginning). As I said
before, it unfortunately does not work.
Note that:
- Marking packets and using policy-based routing is not possible either
(as I understood, the source address has already been computed at this
point and the packet is built, so this is too late)
- Source NATing is also impossible (not implemented on IPv6)
- The /etc/gai.conf tuning file is no help for this purpose either.
I understand this is not a major kernel issue, but this is a really
annoying limitation when you have an almost infinite address space unused :)
[ Note: see also "src attribute ignored for IPv6 (preferred source
address selection)" in linux-netdev mailing-list one year ago. ]
^ permalink raw reply
* Re: [PATCH ] dynamic_debug: call __netdev_printk only for CONFIG_NET
From: Randy Dunlap @ 2011-09-18 17:21 UTC (permalink / raw)
To: Greg KH
Cc: Jason Baron, Arnd Bergmann, gregkh, joe, jim.cromie, bvanassche,
linux-kernel, davem, aloisio.almeida, netdev
In-Reply-To: <20110918082736.GB19444@kroah.com>
On 09/18/2011 01:27 AM, Greg KH wrote:
> On Thu, Sep 01, 2011 at 11:18:18AM -0400, Jason Baron wrote:
>> On Thu, Sep 01, 2011 at 04:57:02PM +0200, Arnd Bergmann wrote:
>>> __netdev_printk is only defined when CONFIG_NET is set. Since we only need
>>> __dynamic_netdev_dbg for network drivers, we can make it conditional on the
>>> same Kconfig symbol.
>>>
>>
>> Hi,
>>
>> Yes, I've posted a fix for this:
>>
>> https://lkml.org/lkml/2011/8/30/297
>>
>> Hopefully, it will be pulled in soon.
>
> As that thread again spun off into confusion, can you please resend the
> end result?
That spinning confusion had nothing to do with the posted & correct patch
which could have been applied several weeks ago.
I'm curious: Do you delete most of your email on a routine basis?
--
~Randy
*** Remember to use Documentation/SubmitChecklist when testing your code ***
^ permalink raw reply
* Re: [GIT] Networking
From: Linus Torvalds @ 2011-09-18 18:06 UTC (permalink / raw)
To: David Miller; +Cc: akpm, netdev, linux-kernel
In-Reply-To: <20110918.022125.1554085675403900813.davem@davemloft.net>
2011/9/17 David Miller <davem@davemloft.net>:
>
> dpward (2):
> net: Make flow cache namespace-aware
> net: Handle different key sizes between address families in flow cache
>
> nhorman (1):
> net: don't clear IFF_XMIT_DST_RELEASE in ether_setup
>
> rajan.aggarwal85@gmail.com (1):
> net/can/af_can.c: Change del_timer to del_timer_sync
Guys, if somebody has such a broken email setup that they don't even
show their own name, don't take patches from them.
If you cannot even set up email sanely, there is zero reason to
believe that the patch should be good. And if the patch is trivial and
you want to take it despite the source of the patch being crap, please
spend the five seconds to fix it up.
Proper names are part of the commit message. Don't make it look like
crap. I get ugly flashbacks to SVN or CVS when I see stuff like this.
Don't do it.
Linus
^ permalink raw reply
* Re: [PATCH 4/4] net/fec: add imx6q enet support
From: Francois Romieu @ 2011-09-18 18:09 UTC (permalink / raw)
To: Shawn Guo; +Cc: netdev, David S. Miller, linux-arm-kernel, patches
In-Reply-To: <1316346852-17090-5-git-send-email-shawn.guo@linaro.org>
Shawn Guo <shawn.guo@linaro.org> :
[...]
> diff --git a/drivers/net/fec.c b/drivers/net/fec.c
> index 04206e4..849cb0b 100644
> --- a/drivers/net/fec.c
> +++ b/drivers/net/fec.c
> @@ -442,18 +453,23 @@ fec_restart(struct net_device *ndev, int duplex)
> /* Enable flow control and length check */
> rcntl |= 0x40000000 | 0x00000020;
>
> - /* MII or RMII */
> + /* RGMII, RMII or MII */
> + if (fep->phy_interface == PHY_INTERFACE_MODE_RGMII)
> + rcntl |= (1 << 6);
> if (fep->phy_interface == PHY_INTERFACE_MODE_RMII)
^^^^ missing "else"
[...]
> + /* 1G, 100M or 10M */
> + if (fep->phy_dev) {
> + if (fep->phy_dev->speed == SPEED_1000)
> + ecntl |= (1 << 8);
> + else if (fep->phy_dev->speed == SPEED_100)
> + rcntl &= ~(1 << 9);
> + else
> + rcntl |= (1 << 9);
> + }
[...]
> + if (id_entry->driver_data & FEC_QUIRK_ENET_MAC) {
> + /* enable ENET endian swap */
> + ecntl |= (1 << 8);
I do not understand why the endian swap bit of ecntl needs to be
set the same in these two different paths, especially as the latter
handles the old faulty imx28 and the former the newly fixed imx6q.
Typo ?
--
Ueimor
^ permalink raw reply
* Re: [PATCH ] dynamic_debug: call __netdev_printk only for CONFIG_NET
From: Greg KH @ 2011-09-18 18:27 UTC (permalink / raw)
To: Randy Dunlap
Cc: Jason Baron, Arnd Bergmann, gregkh, joe, jim.cromie, bvanassche,
linux-kernel, davem, aloisio.almeida, netdev
In-Reply-To: <4E76288A.4020801@xenotime.net>
On Sun, Sep 18, 2011 at 10:21:14AM -0700, Randy Dunlap wrote:
> On 09/18/2011 01:27 AM, Greg KH wrote:
> > On Thu, Sep 01, 2011 at 11:18:18AM -0400, Jason Baron wrote:
> >> On Thu, Sep 01, 2011 at 04:57:02PM +0200, Arnd Bergmann wrote:
> >>> __netdev_printk is only defined when CONFIG_NET is set. Since we only need
> >>> __dynamic_netdev_dbg for network drivers, we can make it conditional on the
> >>> same Kconfig symbol.
> >>>
> >>
> >> Hi,
> >>
> >> Yes, I've posted a fix for this:
> >>
> >> https://lkml.org/lkml/2011/8/30/297
> >>
> >> Hopefully, it will be pulled in soon.
> >
> > As that thread again spun off into confusion, can you please resend the
> > end result?
>
> That spinning confusion had nothing to do with the posted & correct patch
> which could have been applied several weeks ago.
>
> I'm curious: Do you delete most of your email on a routine basis?
No, only after going through pending patches do I purge them. And when
a series of patches generates a thread like this one, where people are
arguing over the way the macros are named, and no one seems to agree, I
will take it as the fact that this series was contentious and needs to
be resent after taking into consideration the original complaints.
For me to keep all email threads, based on the amount of email I get[1],
would be ludicrous.
greg k-h
[1]: http://www.kroah.com/log/linux/get_lots_of_email.html
^ permalink raw reply
* Re: [PATCH ] dynamic_debug: call __netdev_printk only for CONFIG_NET
From: Randy Dunlap @ 2011-09-18 18:42 UTC (permalink / raw)
To: Greg KH
Cc: Jason Baron, Arnd Bergmann, gregkh, joe, jim.cromie, bvanassche,
linux-kernel, davem, aloisio.almeida, netdev
In-Reply-To: <20110918182714.GA32615@kroah.com>
On 09/18/2011 11:27 AM, Greg KH wrote:
> On Sun, Sep 18, 2011 at 10:21:14AM -0700, Randy Dunlap wrote:
>> On 09/18/2011 01:27 AM, Greg KH wrote:
>>> On Thu, Sep 01, 2011 at 11:18:18AM -0400, Jason Baron wrote:
>>>> On Thu, Sep 01, 2011 at 04:57:02PM +0200, Arnd Bergmann wrote:
>>>>> __netdev_printk is only defined when CONFIG_NET is set. Since we only need
>>>>> __dynamic_netdev_dbg for network drivers, we can make it conditional on the
>>>>> same Kconfig symbol.
>>>>>
>>>>
>>>> Hi,
>>>>
>>>> Yes, I've posted a fix for this:
>>>>
>>>> https://lkml.org/lkml/2011/8/30/297
>>>>
>>>> Hopefully, it will be pulled in soon.
>>>
>>> As that thread again spun off into confusion, can you please resend the
>>> end result?
>>
>> That spinning confusion had nothing to do with the posted & correct patch
>> which could have been applied several weeks ago.
>>
>> I'm curious: Do you delete most of your email on a routine basis?
>
> No, only after going through pending patches do I purge them. And when
> a series of patches generates a thread like this one, where people are
> arguing over the way the macros are named, and no one seems to agree, I
> will take it as the fact that this series was contentious and needs to
> be resent after taking into consideration the original complaints.
Sheesh. The naming of the macros has nothing to do with this build fix.
> For me to keep all email threads, based on the amount of email I get[1],
> would be ludicrous.
>
> greg k-h
>
> [1]: http://www.kroah.com/log/linux/get_lots_of_email.html
--
~Randy
*** Remember to use Documentation/SubmitChecklist when testing your code ***
^ permalink raw reply
* Re: [PATCH v2 4/7] per-cgroup tcp buffers control
From: Kirill A. Shutemov @ 2011-09-18 18:58 UTC (permalink / raw)
To: Cyrill Gorcunov
Cc: Glauber Costa, linux-kernel, paul, lizf, kamezawa.hiroyu,
ebiederm, davem, gthelen, netdev, linux-mm
In-Reply-To: <20110917183358.GB2783@moon>
On Sat, Sep 17, 2011 at 10:33:58PM +0400, Cyrill Gorcunov wrote:
> On Sat, Sep 17, 2011 at 09:11:32PM +0300, Kirill A. Shutemov wrote:
> > On Wed, Sep 14, 2011 at 10:46:12PM -0300, Glauber Costa wrote:
> > > +int tcp_init_cgroup_fill(struct proto *prot, struct cgroup *cgrp,
> > > + struct cgroup_subsys *ss)
> > > +{
> > > + prot->enter_memory_pressure = tcp_enter_memory_pressure;
> > > + prot->memory_allocated = memory_allocated_tcp;
> > > + prot->prot_mem = tcp_sysctl_mem;
> > > + prot->sockets_allocated = sockets_allocated_tcp;
> > > + prot->memory_pressure = memory_pressure_tcp;
> >
> > No fancy formatting, please.
> >
>
> What's wrong with having fancy formatting? It's indeed easier to read
> when members are assigned this way. It's always up to maintainer to
> choose what he prefers, but I see nothing wrong in such style (if only it
> doesn't break the style of the whole file).
You have to remove this indenting if you'll reorganize code (e.g. move
part under if(...)).
IMO, it reduces code maintainability.
--
Kirill A. Shutemov
^ permalink raw reply
* Re: [PATCH v2 2/7] socket: initial cgroup code.
From: Kirill A. Shutemov @ 2011-09-18 18:58 UTC (permalink / raw)
To: Glauber Costa
Cc: linux-kernel, paul, lizf, kamezawa.hiroyu, ebiederm, davem,
gthelen, netdev, linux-mm
In-Reply-To: <4E75664B.9070605@parallels.com>
On Sun, Sep 18, 2011 at 12:32:27AM -0300, Glauber Costa wrote:
> On 09/17/2011 02:52 PM, Kirill A. Shutemov wrote:
> > Do we really need to have these functions in the header?
> >
> No, I can move it to memcontrol.c
>
Yes, please.
--
Kirill A. Shutemov
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply
* Re: [PATCH v2 1/7] Basic kernel memory functionality for the Memory Controller
From: Kirill A. Shutemov @ 2011-09-18 19:05 UTC (permalink / raw)
To: Glauber Costa
Cc: linux-kernel, paul, lizf, kamezawa.hiroyu, ebiederm, davem,
gthelen, netdev, linux-mm
In-Reply-To: <4E7567E0.9010401@parallels.com>
On Sun, Sep 18, 2011 at 12:39:12AM -0300, Glauber Costa wrote:
> > No kernel memory accounting for root cgroup, right?
> Not sure. Maybe kernel memory accounting is useful even for root cgroup.
> Same as normal memory accounting... what we want to avoid is kernel
> memory limits. OTOH, if we are not limiting it anyway, accounting it is
> just useless overhead... Even the statistics can then be gathered
> through all
> the proc files that show slab usage, I guess?
It's better to leave root cgroup without accounting. At least for now.
We can add it later if needed.
> >
> >> @@ -3979,6 +3999,10 @@ static u64 mem_cgroup_read(struct cgroup *cont, struct cftype *cft)
> >> else
> >> val = res_counter_read_u64(&mem->memsw, name);
> >> break;
> >> + case _KMEM:
> >> + val = res_counter_read_u64(&mem->kmem, name);
> >> + break;
> >> +
> >
> > Always zero in root cgroup?
>
> Yes, if we're not accounting, it should be zero. WARN_ON, maybe?
-ENOSYS?
--
Kirill A. Shutemov
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply
* Re: [PATCH v2 1/7] Basic kernel memory functionality for the Memory Controller
From: Glauber Costa @ 2011-09-18 19:11 UTC (permalink / raw)
To: Kirill A. Shutemov
Cc: linux-kernel, paul, lizf, kamezawa.hiroyu, ebiederm, davem,
gthelen, netdev, linux-mm
In-Reply-To: <20110918190509.GC28057@shutemov.name>
On 09/18/2011 04:05 PM, Kirill A. Shutemov wrote:
> On Sun, Sep 18, 2011 at 12:39:12AM -0300, Glauber Costa wrote:
>>> No kernel memory accounting for root cgroup, right?
>> Not sure. Maybe kernel memory accounting is useful even for root cgroup.
>> Same as normal memory accounting... what we want to avoid is kernel
>> memory limits. OTOH, if we are not limiting it anyway, accounting it is
>> just useless overhead... Even the statistics can then be gathered
>> through all
>> the proc files that show slab usage, I guess?
>
> It's better to leave root cgroup without accounting. At least for now.
> We can add it later if needed.
Fair.
>>>
>>>> @@ -3979,6 +3999,10 @@ static u64 mem_cgroup_read(struct cgroup *cont, struct cftype *cft)
>>>> else
>>>> val = res_counter_read_u64(&mem->memsw, name);
>>>> break;
>>>> + case _KMEM:
>>>> + val = res_counter_read_u64(&mem->kmem, name);
>>>> + break;
>>>> +
>>>
>>> Always zero in root cgroup?
>>
>> Yes, if we're not accounting, it should be zero. WARN_ON, maybe?
>
> -ENOSYS?
>
I'd personally prefer WARN_ON. It is good symmetry from userspace PoV to
always be able to get a value out of it. Also, it something goes wrong
and it is not zero for some reason, this will help us find it.
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply
* Re: [GIT] Networking
From: Markus Trippelsdorf @ 2011-09-18 19:23 UTC (permalink / raw)
To: Linus Torvalds; +Cc: David Miller, akpm, netdev, linux-kernel, eric.dumazet
In-Reply-To: <CA+55aFxc1tMCXKBV33PVD5BybTLBpfJAEcObvoQwbwqbKOsp7g@mail.gmail.com>
On 2011.09.18 at 11:06 -0700, Linus Torvalds wrote:
> 2011/9/17 David Miller <davem@davemloft.net>:
> >
> > dpward (2):
> > net: Make flow cache namespace-aware
> > net: Handle different key sizes between address families in flow cache
> >
> > nhorman (1):
> > net: don't clear IFF_XMIT_DST_RELEASE in ether_setup
> >
> > rajan.aggarwal85@gmail.com (1):
> > net/can/af_can.c: Change del_timer to del_timer_sync
>
> Guys, if somebody has such a broken email setup that they don't even
> show their own name, don't take patches from them.
>
> If you cannot even set up email sanely, there is zero reason to
> believe that the patch should be good. And if the patch is trivial and
> you want to take it despite the source of the patch being crap, please
> spend the five seconds to fix it up.
>
> Proper names are part of the commit message. Don't make it look like
> crap. I get ugly flashbacks to SVN or CVS when I see stuff like this.
> Don't do it.
Plus commit 946cedccbd73874 breaks the build:
LD init/built-in.o
LD .tmp_vmlinux1
net/built-in.o:sysctl_net.c:function tcp_v4_conn_request: error: undefined reference to 'cookie_v4_init_sequence'
make: *** [.tmp_vmlinux1] Error 1
commit 946cedccbd7387488d2cee5da92cdfeb28d2e670
Author: Eric Dumazet <eric.dumazet@gmail.com>
Date: Tue Aug 30 03:21:44 2011 +0000
tcp: Change possible SYN flooding messages
"Possible SYN flooding on port xxxx " messages can fill logs on servers.
Change logic to log the message only once per listener, and add two new
SNMP counters to track :
TCPReqQFullDoCookies : number of times a SYNCOOKIE was replied to client
TCPReqQFullDrop : number of times a SYN request was dropped because
syncookies were not enabled.
Based on a prior patch from Tom Herbert, and suggestions from David.
--
Markus
^ permalink raw reply
* Re: [BUG?] tcp: potential bug in tcp_is_sackblock_valid()
From: Eric Dumazet @ 2011-09-18 19:35 UTC (permalink / raw)
To: Yan, Zheng
Cc: netdev@vger.kernel.org, davem@davemloft.net, herbert,
sfr@canb.auug.org.au
In-Reply-To: <4E696FD0.7060702@intel.com>
Le vendredi 09 septembre 2011 à 09:45 +0800, Yan, Zheng a écrit :
> Hi all,
>
> I found a check in tcp_is_sackblock_valid() is suspicious. It against
> its comment and RFC. I think the correct check should be:
>
> ---
> diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
> index 385c470..a5d01b1 100644
> --- a/net/ipv4/tcp_input.c
> +++ b/net/ipv4/tcp_input.c
> @@ -1124,7 +1124,7 @@ static int tcp_is_sackblock_valid(struct tcp_sock *tp, int is_dsack,
> return 0;
>
> /* ...Then it's D-SACK, and must reside below snd_una completely */
> - if (!after(end_seq, tp->snd_una))
> + if (after(end_seq, tp->snd_una))
> return 0;
>
> if (!before(start_seq, tp->undo_marker))
> ---
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
This bug was introduced in 2.6.24 by commit 5b3c9882
^ permalink raw reply
* Re: [PATCH v2 4/7] per-cgroup tcp buffers control
From: Glauber Costa @ 2011-09-18 19:42 UTC (permalink / raw)
To: Kirill A. Shutemov
Cc: Cyrill Gorcunov, linux-kernel, paul, lizf, kamezawa.hiroyu,
ebiederm, davem, gthelen, netdev, linux-mm
In-Reply-To: <20110918185806.GA28057@shutemov.name>
On 09/18/2011 03:58 PM, Kirill A. Shutemov wrote:
> On Sat, Sep 17, 2011 at 10:33:58PM +0400, Cyrill Gorcunov wrote:
>> On Sat, Sep 17, 2011 at 09:11:32PM +0300, Kirill A. Shutemov wrote:
>>> On Wed, Sep 14, 2011 at 10:46:12PM -0300, Glauber Costa wrote:
>>>> +int tcp_init_cgroup_fill(struct proto *prot, struct cgroup *cgrp,
>>>> + struct cgroup_subsys *ss)
>>>> +{
>>>> + prot->enter_memory_pressure = tcp_enter_memory_pressure;
>>>> + prot->memory_allocated = memory_allocated_tcp;
>>>> + prot->prot_mem = tcp_sysctl_mem;
>>>> + prot->sockets_allocated = sockets_allocated_tcp;
>>>> + prot->memory_pressure = memory_pressure_tcp;
>>>
>>> No fancy formatting, please.
>>>
>>
>> What's wrong with having fancy formatting? It's indeed easier to read
>> when members are assigned this way. It's always up to maintainer to
>> choose what he prefers, but I see nothing wrong in such style (if only it
>> doesn't break the style of the whole file).
>
> You have to remove this indenting if you'll reorganize code (e.g. move
> part under if(...)).
> IMO, it reduces code maintainability.
>
As I said, I don't care, so I'll change. But I have to say I disagree
with your statement.
It is a pack of assignments, so if you reorganize this code, two things
can happen:
1) It is not moved to a new ident level -> It keeps being a pack of
assignments, and you don't really need to change it.
2) It is moved to a new ident level -> You have to touch it anyway...
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply
* Re: [GIT] Networking
From: Eric Dumazet @ 2011-09-18 19:46 UTC (permalink / raw)
To: Markus Trippelsdorf
Cc: Linus Torvalds, David Miller, akpm, netdev, linux-kernel
In-Reply-To: <20110918192333.GA1641@x4.trippels.de>
Le dimanche 18 septembre 2011 à 21:23 +0200, Markus Trippelsdorf a
écrit :
> On 2011.09.18 at 11:06 -0700, Linus Torvalds wrote:
> > 2011/9/17 David Miller <davem@davemloft.net>:
> > >
> > > dpward (2):
> > > net: Make flow cache namespace-aware
> > > net: Handle different key sizes between address families in flow cache
> > >
> > > nhorman (1):
> > > net: don't clear IFF_XMIT_DST_RELEASE in ether_setup
> > >
> > > rajan.aggarwal85@gmail.com (1):
> > > net/can/af_can.c: Change del_timer to del_timer_sync
> >
> > Guys, if somebody has such a broken email setup that they don't even
> > show their own name, don't take patches from them.
> >
> > If you cannot even set up email sanely, there is zero reason to
> > believe that the patch should be good. And if the patch is trivial and
> > you want to take it despite the source of the patch being crap, please
> > spend the five seconds to fix it up.
> >
> > Proper names are part of the commit message. Don't make it look like
> > crap. I get ugly flashbacks to SVN or CVS when I see stuff like this.
> > Don't do it.
>
> Plus commit 946cedccbd73874 breaks the build:
>
> LD init/built-in.o
> LD .tmp_vmlinux1
> net/built-in.o:sysctl_net.c:function tcp_v4_conn_request: error: undefined reference to 'cookie_v4_init_sequence'
> make: *** [.tmp_vmlinux1] Error 1
>
> commit 946cedccbd7387488d2cee5da92cdfeb28d2e670
> Author: Eric Dumazet <eric.dumazet@gmail.com>
> Date: Tue Aug 30 03:21:44 2011 +0000
>
> tcp: Change possible SYN flooding messages
>
> "Possible SYN flooding on port xxxx " messages can fill logs on servers.
>
> Change logic to log the message only once per listener, and add two new
> SNMP counters to track :
>
> TCPReqQFullDoCookies : number of times a SYNCOOKIE was replied to client
>
> TCPReqQFullDrop : number of times a SYN request was dropped because
> syncookies were not enabled.
>
> Based on a prior patch from Tom Herbert, and suggestions from David.
>
>
Oh well, trying to remove those ugly #ifdef was not so easy.
I'll cook a patch, thanks for the report
^ permalink raw reply
* Re: [GIT] Networking
From: Markus Trippelsdorf @ 2011-09-18 19:48 UTC (permalink / raw)
To: Eric Dumazet; +Cc: Linus Torvalds, David Miller, akpm, netdev, linux-kernel
In-Reply-To: <1316375164.31335.18.camel@edumazet-laptop>
On 2011.09.18 at 21:46 +0200, Eric Dumazet wrote:
> Le dimanche 18 septembre 2011 à 21:23 +0200, Markus Trippelsdorf a
> écrit :
> > On 2011.09.18 at 11:06 -0700, Linus Torvalds wrote:
> > > 2011/9/17 David Miller <davem@davemloft.net>:
> > > >
> > > > dpward (2):
> > > > net: Make flow cache namespace-aware
> > > > net: Handle different key sizes between address families in flow cache
> > > >
> > > > nhorman (1):
> > > > net: don't clear IFF_XMIT_DST_RELEASE in ether_setup
> > > >
> > > > rajan.aggarwal85@gmail.com (1):
> > > > net/can/af_can.c: Change del_timer to del_timer_sync
> > >
> > > Guys, if somebody has such a broken email setup that they don't even
> > > show their own name, don't take patches from them.
> > >
> > > If you cannot even set up email sanely, there is zero reason to
> > > believe that the patch should be good. And if the patch is trivial and
> > > you want to take it despite the source of the patch being crap, please
> > > spend the five seconds to fix it up.
> > >
> > > Proper names are part of the commit message. Don't make it look like
> > > crap. I get ugly flashbacks to SVN or CVS when I see stuff like this.
> > > Don't do it.
> >
> > Plus commit 946cedccbd73874 breaks the build:
> >
> > LD init/built-in.o
> > LD .tmp_vmlinux1
> > net/built-in.o:sysctl_net.c:function tcp_v4_conn_request: error: undefined reference to 'cookie_v4_init_sequence'
> > make: *** [.tmp_vmlinux1] Error 1
> >
> > commit 946cedccbd7387488d2cee5da92cdfeb28d2e670
> > Author: Eric Dumazet <eric.dumazet@gmail.com>
> > Date: Tue Aug 30 03:21:44 2011 +0000
> >
> > tcp: Change possible SYN flooding messages
> >
> > "Possible SYN flooding on port xxxx " messages can fill logs on servers.
> >
> > Change logic to log the message only once per listener, and add two new
> > SNMP counters to track :
> >
> > TCPReqQFullDoCookies : number of times a SYNCOOKIE was replied to client
> >
> > TCPReqQFullDrop : number of times a SYN request was dropped because
> > syncookies were not enabled.
> >
> > Based on a prior patch from Tom Herbert, and suggestions from David.
> >
> >
>
> Oh well, trying to remove those ugly #ifdef was not so easy.
> I'll cook a patch, thanks for the report
The following works for me:
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index c34f015..ef9dd55 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -1264,7 +1264,9 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb)
* evidently real one.
*/
if (inet_csk_reqsk_queue_is_full(sk) && !isn) {
+#ifdef CONFIG_SYN_COOKIES
want_cookie = tcp_syn_flood_action(sk, skb, "TCP");
+#endif
if (!want_cookie)
goto drop;
}
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 3c9fa61..7ffc3b1 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -1174,7 +1174,9 @@ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb)
goto drop;
if (inet_csk_reqsk_queue_is_full(sk) && !isn) {
+#ifdef CONFIG_SYN_COOKIES
want_cookie = tcp_syn_flood_action(sk, skb, "TCPv6");
+#endif
if (!want_cookie)
goto drop;
}
--
Markus
^ 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