* pull request: wireless-2.6 2010-11-10
From: John W. Linville @ 2010-11-10 18:58 UTC (permalink / raw)
To: davem; +Cc: linux-wireless, netdev, linux-kernel
David,
Here is a batch of fixes intended for 2.6.37. Also mixed-in are a few
device ID updates. I think the changelog entries are reasonably
documentary of the issues being fixed, so I won't belabor them. ;-)
This also includes a round of Bluetooth fixes from Gustavo:
"The following batch contains some bugfixes for 2.6.37. A fix for unaligned
access in L2CAP, a Kconfig error, two fixes related to security of
the Bluetooth links, and support for the a MacBook Air Bluetooth device.
There is also a one line patch from Matthew Garret, that enables USB
autosuspend for btusb module, which shall be completely safe. Please
pull, thanks."
Please let me know if there are problems!
Thanks,
John
---
The following changes since commit 88f8a5e3e7defccd3925cabb1ee4d3994e5cdb52:
net: tipc: fix information leak to userland (2010-11-09 09:25:46 -0800)
are available in the git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/linville/wireless-2.6.git master
Brian Cavagnolo (1):
mac80211: unset SDATA_STATE_OFFCHANNEL when cancelling a scan
Christian Lamparter (1):
carl9170: usbid table updates
Daniel Drake (1):
libertas: terminate scan when stopping interface
Edgar (gimli) Hucek (1):
Bluetooth: Add MacBookAir3,1(2) support
Felix Fietkau (2):
cfg80211: fix a crash in dev lookup on dump commands
ath9k: check old power mode before clearing cycle counters
Gustavo F. Padovan (1):
Bluetooth: fix endianness conversion in L2CAP
Haitao Zhang (1):
ath9k_htc: Add support for device ID 3346
Johan Hedberg (1):
Bluetooth: Fix non-SSP auth request for HIGH security level sockets
Linus Torvalds (1):
libipw: fix proc entry removal
Luiz Augusto von Dentz (1):
Bluetooth: fix not setting security level when creating a rfcomm session
Matthew Garrett (1):
Bluetooth: Enable USB autosuspend by default on btusb
Rajkumar Manoharan (3):
ath9k: Avoid HW opmode overridden on monitor mode changes
ath9k_htc: Fix probe failure if CONFIG_USB_DEBUG enabled
ath9k_hw: Fix memory leak on ath9k_hw_rf_alloc_ext_banks failure
Randy Dunlap (1):
Bluetooth: fix hidp kconfig dependency warning
Vasanthakumar Thiagarajan (1):
ath9k_hw: Fix AR9280 surprise removal during frequent idle on/off
Vivek Natarajan (1):
ath9k: Fix a DMA latency issue for Intel Pinetrail platforms.
Wey-Yi Guy (1):
iwlwifi: dont use pci_dev before it being assign
steven miao (1):
Bluetooth: fix unaligned access to l2cap conf data
drivers/bluetooth/btusb.c | 5 ++++
drivers/net/wireless/ath/ath9k/ar9002_hw.c | 3 ++
drivers/net/wireless/ath/ath9k/ath9k.h | 1 +
drivers/net/wireless/ath/ath9k/hif_usb.c | 31 ++++++++++++-------------
drivers/net/wireless/ath/ath9k/hw.c | 15 +++++++++++-
drivers/net/wireless/ath/ath9k/hw.h | 1 +
drivers/net/wireless/ath/ath9k/init.c | 8 ++++++
drivers/net/wireless/ath/ath9k/main.c | 29 +++++++++++++++++------
drivers/net/wireless/ath/ath9k/recv.c | 4 +-
drivers/net/wireless/ath/ath9k/reg.h | 1 +
drivers/net/wireless/ath/carl9170/usb.c | 4 ++-
drivers/net/wireless/ipw2x00/libipw_module.c | 9 ++++---
drivers/net/wireless/iwlwifi/iwl3945-base.c | 3 +-
drivers/net/wireless/libertas/cfg.c | 5 ++-
drivers/net/wireless/libertas/dev.h | 1 +
drivers/net/wireless/libertas/main.c | 7 ++++++
net/bluetooth/hci_event.c | 6 +++++
net/bluetooth/hidp/Kconfig | 2 +-
net/bluetooth/l2cap.c | 8 +++---
net/bluetooth/rfcomm/core.c | 13 ++++++++--
net/mac80211/iface.c | 6 ++--
net/wireless/nl80211.c | 4 +-
22 files changed, 117 insertions(+), 49 deletions(-)
diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c
index d120a5c..ab3894f 100644
--- a/drivers/bluetooth/btusb.c
+++ b/drivers/bluetooth/btusb.c
@@ -68,6 +68,9 @@ static struct usb_device_id btusb_table[] = {
/* Apple MacBookPro6,2 */
{ USB_DEVICE(0x05ac, 0x8218) },
+ /* Apple MacBookAir3,1, MacBookAir3,2 */
+ { USB_DEVICE(0x05ac, 0x821b) },
+
/* AVM BlueFRITZ! USB v2.0 */
{ USB_DEVICE(0x057c, 0x3800) },
@@ -1029,6 +1032,8 @@ static int btusb_probe(struct usb_interface *intf,
usb_set_intfdata(intf, data);
+ usb_enable_autosuspend(interface_to_usbdev(intf));
+
return 0;
}
diff --git a/drivers/net/wireless/ath/ath9k/ar9002_hw.c b/drivers/net/wireless/ath/ath9k/ar9002_hw.c
index a0471f2..48261b7 100644
--- a/drivers/net/wireless/ath/ath9k/ar9002_hw.c
+++ b/drivers/net/wireless/ath/ath9k/ar9002_hw.c
@@ -410,6 +410,9 @@ static void ar9002_hw_configpcipowersave(struct ath_hw *ah,
val &= ~(AR_WA_BIT6 | AR_WA_BIT7);
}
+ if (AR_SREV_9280(ah))
+ val |= AR_WA_BIT22;
+
if (AR_SREV_9285E_20(ah))
val |= AR_WA_BIT23;
diff --git a/drivers/net/wireless/ath/ath9k/ath9k.h b/drivers/net/wireless/ath/ath9k/ath9k.h
index 9b8e7e3..170d44a 100644
--- a/drivers/net/wireless/ath/ath9k/ath9k.h
+++ b/drivers/net/wireless/ath/ath9k/ath9k.h
@@ -675,6 +675,7 @@ static inline void ath_read_cachesize(struct ath_common *common, int *csz)
}
extern struct ieee80211_ops ath9k_ops;
+extern struct pm_qos_request_list ath9k_pm_qos_req;
extern int modparam_nohwcrypt;
extern int led_blink;
diff --git a/drivers/net/wireless/ath/ath9k/hif_usb.c b/drivers/net/wireless/ath/ath9k/hif_usb.c
index 6576f68..f7ec31b 100644
--- a/drivers/net/wireless/ath/ath9k/hif_usb.c
+++ b/drivers/net/wireless/ath/ath9k/hif_usb.c
@@ -35,6 +35,7 @@ static struct usb_device_id ath9k_hif_usb_ids[] = {
{ USB_DEVICE(0x07D1, 0x3A10) }, /* Dlink Wireless 150 */
{ USB_DEVICE(0x13D3, 0x3327) }, /* Azurewave */
{ USB_DEVICE(0x13D3, 0x3328) }, /* Azurewave */
+ { USB_DEVICE(0x13D3, 0x3346) }, /* IMC Networks */
{ USB_DEVICE(0x04CA, 0x4605) }, /* Liteon */
{ USB_DEVICE(0x083A, 0xA704) }, /* SMC Networks */
{ },
@@ -540,11 +541,11 @@ static void ath9k_hif_usb_reg_in_cb(struct urb *urb)
return;
}
- usb_fill_int_urb(urb, hif_dev->udev,
+ usb_fill_bulk_urb(urb, hif_dev->udev,
usb_rcvbulkpipe(hif_dev->udev,
USB_REG_IN_PIPE),
nskb->data, MAX_REG_IN_BUF_SIZE,
- ath9k_hif_usb_reg_in_cb, nskb, 1);
+ ath9k_hif_usb_reg_in_cb, nskb);
ret = usb_submit_urb(urb, GFP_ATOMIC);
if (ret) {
@@ -720,11 +721,11 @@ static int ath9k_hif_usb_alloc_reg_in_urb(struct hif_device_usb *hif_dev)
if (!skb)
goto err;
- usb_fill_int_urb(hif_dev->reg_in_urb, hif_dev->udev,
+ usb_fill_bulk_urb(hif_dev->reg_in_urb, hif_dev->udev,
usb_rcvbulkpipe(hif_dev->udev,
USB_REG_IN_PIPE),
skb->data, MAX_REG_IN_BUF_SIZE,
- ath9k_hif_usb_reg_in_cb, skb, 1);
+ ath9k_hif_usb_reg_in_cb, skb);
if (usb_submit_urb(hif_dev->reg_in_urb, GFP_KERNEL) != 0)
goto err;
@@ -843,14 +844,6 @@ static int ath9k_hif_usb_dev_init(struct hif_device_usb *hif_dev)
goto err_fw_req;
}
- /* Alloc URBs */
- ret = ath9k_hif_usb_alloc_urbs(hif_dev);
- if (ret) {
- dev_err(&hif_dev->udev->dev,
- "ath9k_htc: Unable to allocate URBs\n");
- goto err_urb;
- }
-
/* Download firmware */
ret = ath9k_hif_usb_download_fw(hif_dev);
if (ret) {
@@ -866,16 +859,22 @@ static int ath9k_hif_usb_dev_init(struct hif_device_usb *hif_dev)
*/
for (idx = 0; idx < alt->desc.bNumEndpoints; idx++) {
endp = &alt->endpoint[idx].desc;
- if (((endp->bEndpointAddress & USB_ENDPOINT_NUMBER_MASK)
- == 0x04) &&
- ((endp->bmAttributes & USB_ENDPOINT_XFERTYPE_MASK)
- == USB_ENDPOINT_XFER_INT)) {
+ if ((endp->bmAttributes & USB_ENDPOINT_XFERTYPE_MASK)
+ == USB_ENDPOINT_XFER_INT) {
endp->bmAttributes &= ~USB_ENDPOINT_XFERTYPE_MASK;
endp->bmAttributes |= USB_ENDPOINT_XFER_BULK;
endp->bInterval = 0;
}
}
+ /* Alloc URBs */
+ ret = ath9k_hif_usb_alloc_urbs(hif_dev);
+ if (ret) {
+ dev_err(&hif_dev->udev->dev,
+ "ath9k_htc: Unable to allocate URBs\n");
+ goto err_urb;
+ }
+
return 0;
err_fw_download:
diff --git a/drivers/net/wireless/ath/ath9k/hw.c b/drivers/net/wireless/ath/ath9k/hw.c
index cc13ee1..6ebc68b 100644
--- a/drivers/net/wireless/ath/ath9k/hw.c
+++ b/drivers/net/wireless/ath/ath9k/hw.c
@@ -484,6 +484,7 @@ static int ath9k_hw_post_init(struct ath_hw *ah)
ath_print(ath9k_hw_common(ah), ATH_DBG_FATAL,
"Failed allocating banks for "
"external radio\n");
+ ath9k_hw_rf_free_ext_banks(ah);
return ecode;
}
@@ -952,9 +953,12 @@ static void ath9k_hw_set_operating_mode(struct ath_hw *ah, int opmode)
REG_SET_BIT(ah, AR_CFG, AR_CFG_AP_ADHOC_INDICATION);
break;
case NL80211_IFTYPE_STATION:
- case NL80211_IFTYPE_MONITOR:
REG_WRITE(ah, AR_STA_ID1, val | AR_STA_ID1_KSRCH_MODE);
break;
+ default:
+ if (ah->is_monitoring)
+ REG_WRITE(ah, AR_STA_ID1, val | AR_STA_ID1_KSRCH_MODE);
+ break;
}
}
@@ -1634,7 +1638,6 @@ void ath9k_hw_beaconinit(struct ath_hw *ah, u32 next_beacon, u32 beacon_period)
switch (ah->opmode) {
case NL80211_IFTYPE_STATION:
- case NL80211_IFTYPE_MONITOR:
REG_WRITE(ah, AR_NEXT_TBTT_TIMER, TU_TO_USEC(next_beacon));
REG_WRITE(ah, AR_NEXT_DMA_BEACON_ALERT, 0xffff);
REG_WRITE(ah, AR_NEXT_SWBA, 0x7ffff);
@@ -1663,6 +1666,14 @@ void ath9k_hw_beaconinit(struct ath_hw *ah, u32 next_beacon, u32 beacon_period)
AR_TBTT_TIMER_EN | AR_DBA_TIMER_EN | AR_SWBA_TIMER_EN;
break;
default:
+ if (ah->is_monitoring) {
+ REG_WRITE(ah, AR_NEXT_TBTT_TIMER,
+ TU_TO_USEC(next_beacon));
+ REG_WRITE(ah, AR_NEXT_DMA_BEACON_ALERT, 0xffff);
+ REG_WRITE(ah, AR_NEXT_SWBA, 0x7ffff);
+ flags |= AR_TBTT_TIMER_EN;
+ break;
+ }
ath_print(ath9k_hw_common(ah), ATH_DBG_BEACON,
"%s: unsupported opmode: %d\n",
__func__, ah->opmode);
diff --git a/drivers/net/wireless/ath/ath9k/hw.h b/drivers/net/wireless/ath/ath9k/hw.h
index d032939..d47d1b4 100644
--- a/drivers/net/wireless/ath/ath9k/hw.h
+++ b/drivers/net/wireless/ath/ath9k/hw.h
@@ -622,6 +622,7 @@ struct ath_hw {
bool sw_mgmt_crypto;
bool is_pciexpress;
+ bool is_monitoring;
bool need_an_top2_fixup;
u16 tx_trig_level;
diff --git a/drivers/net/wireless/ath/ath9k/init.c b/drivers/net/wireless/ath/ath9k/init.c
index 95b41db..6a0d99e 100644
--- a/drivers/net/wireless/ath/ath9k/init.c
+++ b/drivers/net/wireless/ath/ath9k/init.c
@@ -15,6 +15,7 @@
*/
#include <linux/slab.h>
+#include <linux/pm_qos_params.h>
#include "ath9k.h"
@@ -179,6 +180,8 @@ static const struct ath_ops ath9k_common_ops = {
.write = ath9k_iowrite32,
};
+struct pm_qos_request_list ath9k_pm_qos_req;
+
/**************************/
/* Initialization */
/**************************/
@@ -756,6 +759,9 @@ int ath9k_init_device(u16 devid, struct ath_softc *sc, u16 subsysid,
ath_init_leds(sc);
ath_start_rfkill_poll(sc);
+ pm_qos_add_request(&ath9k_pm_qos_req, PM_QOS_CPU_DMA_LATENCY,
+ PM_QOS_DEFAULT_VALUE);
+
return 0;
error_world:
@@ -811,6 +817,8 @@ void ath9k_deinit_device(struct ath_softc *sc)
ath9k_ps_wakeup(sc);
+ pm_qos_remove_request(&ath9k_pm_qos_req);
+
wiphy_rfkill_stop_polling(sc->hw->wiphy);
ath_deinit_leds(sc);
diff --git a/drivers/net/wireless/ath/ath9k/main.c b/drivers/net/wireless/ath/ath9k/main.c
index b52f1cf..25d3ef4 100644
--- a/drivers/net/wireless/ath/ath9k/main.c
+++ b/drivers/net/wireless/ath/ath9k/main.c
@@ -15,6 +15,7 @@
*/
#include <linux/nl80211.h>
+#include <linux/pm_qos_params.h>
#include "ath9k.h"
#include "btcoex.h"
@@ -93,11 +94,13 @@ void ath9k_ps_wakeup(struct ath_softc *sc)
{
struct ath_common *common = ath9k_hw_common(sc->sc_ah);
unsigned long flags;
+ enum ath9k_power_mode power_mode;
spin_lock_irqsave(&sc->sc_pm_lock, flags);
if (++sc->ps_usecount != 1)
goto unlock;
+ power_mode = sc->sc_ah->power_mode;
ath9k_hw_setpower(sc->sc_ah, ATH9K_PM_AWAKE);
/*
@@ -105,10 +108,12 @@ void ath9k_ps_wakeup(struct ath_softc *sc)
* useful data. Better clear them now so that they don't mess up
* survey data results.
*/
- spin_lock(&common->cc_lock);
- ath_hw_cycle_counters_update(common);
- memset(&common->cc_survey, 0, sizeof(common->cc_survey));
- spin_unlock(&common->cc_lock);
+ if (power_mode != ATH9K_PM_AWAKE) {
+ spin_lock(&common->cc_lock);
+ ath_hw_cycle_counters_update(common);
+ memset(&common->cc_survey, 0, sizeof(common->cc_survey));
+ spin_unlock(&common->cc_lock);
+ }
unlock:
spin_unlock_irqrestore(&sc->sc_pm_lock, flags);
@@ -1217,6 +1222,7 @@ static int ath9k_start(struct ieee80211_hw *hw)
ah->imask |= ATH9K_INT_CST;
sc->sc_flags &= ~SC_OP_INVALID;
+ sc->sc_ah->is_monitoring = false;
/* Disable BMISS interrupt when we're not associated */
ah->imask &= ~(ATH9K_INT_SWBA | ATH9K_INT_BMISS);
@@ -1238,6 +1244,8 @@ static int ath9k_start(struct ieee80211_hw *hw)
ath9k_btcoex_timer_resume(sc);
}
+ pm_qos_update_request(&ath9k_pm_qos_req, 55);
+
mutex_unlock:
mutex_unlock(&sc->mutex);
@@ -1415,6 +1423,8 @@ static void ath9k_stop(struct ieee80211_hw *hw)
sc->sc_flags |= SC_OP_INVALID;
+ pm_qos_update_request(&ath9k_pm_qos_req, PM_QOS_DEFAULT_VALUE);
+
mutex_unlock(&sc->mutex);
ath_print(common, ATH_DBG_CONFIG, "Driver halt\n");
@@ -1493,8 +1503,7 @@ static int ath9k_add_interface(struct ieee80211_hw *hw,
ath9k_hw_set_interrupts(ah, ah->imask);
if (vif->type == NL80211_IFTYPE_AP ||
- vif->type == NL80211_IFTYPE_ADHOC ||
- vif->type == NL80211_IFTYPE_MONITOR) {
+ vif->type == NL80211_IFTYPE_ADHOC) {
sc->sc_flags |= SC_OP_ANI_RUN;
ath_start_ani(common);
}
@@ -1644,8 +1653,12 @@ static int ath9k_config(struct ieee80211_hw *hw, u32 changed)
if (changed & IEEE80211_CONF_CHANGE_MONITOR) {
if (conf->flags & IEEE80211_CONF_MONITOR) {
ath_print(common, ATH_DBG_CONFIG,
- "HW opmode set to Monitor mode\n");
- sc->sc_ah->opmode = NL80211_IFTYPE_MONITOR;
+ "Monitor mode is enabled\n");
+ sc->sc_ah->is_monitoring = true;
+ } else {
+ ath_print(common, ATH_DBG_CONFIG,
+ "Monitor mode is disabled\n");
+ sc->sc_ah->is_monitoring = false;
}
}
diff --git a/drivers/net/wireless/ath/ath9k/recv.c b/drivers/net/wireless/ath/ath9k/recv.c
index fddb012..c76ea53 100644
--- a/drivers/net/wireless/ath/ath9k/recv.c
+++ b/drivers/net/wireless/ath/ath9k/recv.c
@@ -441,7 +441,7 @@ u32 ath_calcrxfilter(struct ath_softc *sc)
*/
if (((sc->sc_ah->opmode != NL80211_IFTYPE_AP) &&
(sc->rx.rxfilter & FIF_PROMISC_IN_BSS)) ||
- (sc->sc_ah->opmode == NL80211_IFTYPE_MONITOR))
+ (sc->sc_ah->is_monitoring))
rfilt |= ATH9K_RX_FILTER_PROM;
if (sc->rx.rxfilter & FIF_CONTROL)
@@ -897,7 +897,7 @@ static bool ath9k_rx_accept(struct ath_common *common,
* decryption and MIC failures. For monitor mode,
* we also ignore the CRC error.
*/
- if (ah->opmode == NL80211_IFTYPE_MONITOR) {
+ if (ah->is_monitoring) {
if (rx_stats->rs_status &
~(ATH9K_RXERR_DECRYPT | ATH9K_RXERR_MIC |
ATH9K_RXERR_CRC))
diff --git a/drivers/net/wireless/ath/ath9k/reg.h b/drivers/net/wireless/ath/ath9k/reg.h
index 42976b0..fa05b71 100644
--- a/drivers/net/wireless/ath/ath9k/reg.h
+++ b/drivers/net/wireless/ath/ath9k/reg.h
@@ -703,6 +703,7 @@
#define AR_WA_RESET_EN (1 << 18) /* Sw Control to enable PCI-Reset to POR (bit 15) */
#define AR_WA_ANALOG_SHIFT (1 << 20)
#define AR_WA_POR_SHORT (1 << 21) /* PCI-E Phy reset control */
+#define AR_WA_BIT22 (1 << 22)
#define AR9285_WA_DEFAULT 0x004a050b
#define AR9280_WA_DEFAULT 0x0040073b
#define AR_WA_DEFAULT 0x0000073f
diff --git a/drivers/net/wireless/ath/carl9170/usb.c b/drivers/net/wireless/ath/carl9170/usb.c
index d8607f4..3317039 100644
--- a/drivers/net/wireless/ath/carl9170/usb.c
+++ b/drivers/net/wireless/ath/carl9170/usb.c
@@ -82,9 +82,11 @@ static struct usb_device_id carl9170_usb_ids[] = {
{ USB_DEVICE(0x07d1, 0x3c10) },
/* D-Link DWA 160 A2 */
{ USB_DEVICE(0x07d1, 0x3a09) },
+ /* D-Link DWA 130 D */
+ { USB_DEVICE(0x07d1, 0x3a0f) },
/* Netgear WNA1000 */
{ USB_DEVICE(0x0846, 0x9040) },
- /* Netgear WNDA3100 */
+ /* Netgear WNDA3100 (v1) */
{ USB_DEVICE(0x0846, 0x9010) },
/* Netgear WN111 v2 */
{ USB_DEVICE(0x0846, 0x9001), .driver_info = CARL9170_ONE_LED },
diff --git a/drivers/net/wireless/ipw2x00/libipw_module.c b/drivers/net/wireless/ipw2x00/libipw_module.c
index 32dee2c..d5ef696 100644
--- a/drivers/net/wireless/ipw2x00/libipw_module.c
+++ b/drivers/net/wireless/ipw2x00/libipw_module.c
@@ -54,6 +54,7 @@
#define DRV_DESCRIPTION "802.11 data/management/control stack"
#define DRV_NAME "libipw"
+#define DRV_PROCNAME "ieee80211"
#define DRV_VERSION LIBIPW_VERSION
#define DRV_COPYRIGHT "Copyright (C) 2004-2005 Intel Corporation <jketreno@linux.intel.com>"
@@ -293,16 +294,16 @@ static int __init libipw_init(void)
struct proc_dir_entry *e;
libipw_debug_level = debug;
- libipw_proc = proc_mkdir("ieee80211", init_net.proc_net);
+ libipw_proc = proc_mkdir(DRV_PROCNAME, init_net.proc_net);
if (libipw_proc == NULL) {
- LIBIPW_ERROR("Unable to create " DRV_NAME
+ LIBIPW_ERROR("Unable to create " DRV_PROCNAME
" proc directory\n");
return -EIO;
}
e = proc_create("debug_level", S_IRUGO | S_IWUSR, libipw_proc,
&debug_level_proc_fops);
if (!e) {
- remove_proc_entry(DRV_NAME, init_net.proc_net);
+ remove_proc_entry(DRV_PROCNAME, init_net.proc_net);
libipw_proc = NULL;
return -EIO;
}
@@ -319,7 +320,7 @@ static void __exit libipw_exit(void)
#ifdef CONFIG_LIBIPW_DEBUG
if (libipw_proc) {
remove_proc_entry("debug_level", libipw_proc);
- remove_proc_entry(DRV_NAME, init_net.proc_net);
+ remove_proc_entry(DRV_PROCNAME, init_net.proc_net);
libipw_proc = NULL;
}
#endif /* CONFIG_LIBIPW_DEBUG */
diff --git a/drivers/net/wireless/iwlwifi/iwl3945-base.c b/drivers/net/wireless/iwlwifi/iwl3945-base.c
index 8f8c4b7..7edf8c2 100644
--- a/drivers/net/wireless/iwlwifi/iwl3945-base.c
+++ b/drivers/net/wireless/iwlwifi/iwl3945-base.c
@@ -4000,7 +4000,8 @@ static int iwl3945_pci_probe(struct pci_dev *pdev, const struct pci_device_id *e
* "the hard way", rather than using device's scan.
*/
if (iwl3945_mod_params.disable_hw_scan) {
- IWL_ERR(priv, "sw scan support is deprecated\n");
+ dev_printk(KERN_DEBUG, &(pdev->dev),
+ "sw scan support is deprecated\n");
iwl3945_hw_ops.hw_scan = NULL;
}
diff --git a/drivers/net/wireless/libertas/cfg.c b/drivers/net/wireless/libertas/cfg.c
index 5046a00..373930a 100644
--- a/drivers/net/wireless/libertas/cfg.c
+++ b/drivers/net/wireless/libertas/cfg.c
@@ -700,8 +700,9 @@ static void lbs_scan_worker(struct work_struct *work)
if (priv->scan_channel < priv->scan_req->n_channels) {
cancel_delayed_work(&priv->scan_work);
- queue_delayed_work(priv->work_thread, &priv->scan_work,
- msecs_to_jiffies(300));
+ if (!priv->stopping)
+ queue_delayed_work(priv->work_thread, &priv->scan_work,
+ msecs_to_jiffies(300));
}
/* This is the final data we are about to send */
diff --git a/drivers/net/wireless/libertas/dev.h b/drivers/net/wireless/libertas/dev.h
index f062ed5..cb14c38 100644
--- a/drivers/net/wireless/libertas/dev.h
+++ b/drivers/net/wireless/libertas/dev.h
@@ -36,6 +36,7 @@ struct lbs_private {
/* CFG80211 */
struct wireless_dev *wdev;
bool wiphy_registered;
+ bool stopping;
struct cfg80211_scan_request *scan_req;
u8 assoc_bss[ETH_ALEN];
u8 disassoc_reason;
diff --git a/drivers/net/wireless/libertas/main.c b/drivers/net/wireless/libertas/main.c
index 47ce5a6..46b88b1 100644
--- a/drivers/net/wireless/libertas/main.c
+++ b/drivers/net/wireless/libertas/main.c
@@ -104,6 +104,7 @@ static int lbs_dev_open(struct net_device *dev)
lbs_deb_enter(LBS_DEB_NET);
spin_lock_irq(&priv->driver_lock);
+ priv->stopping = false;
if (priv->connect_status == LBS_CONNECTED)
netif_carrier_on(dev);
@@ -131,10 +132,16 @@ static int lbs_eth_stop(struct net_device *dev)
lbs_deb_enter(LBS_DEB_NET);
spin_lock_irq(&priv->driver_lock);
+ priv->stopping = true;
netif_stop_queue(dev);
spin_unlock_irq(&priv->driver_lock);
schedule_work(&priv->mcast_work);
+ cancel_delayed_work_sync(&priv->scan_work);
+ if (priv->scan_req) {
+ cfg80211_scan_done(priv->scan_req, false);
+ priv->scan_req = NULL;
+ }
lbs_deb_leave(LBS_DEB_NET);
return 0;
diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c
index bfef5ba..84093b0 100644
--- a/net/bluetooth/hci_event.c
+++ b/net/bluetooth/hci_event.c
@@ -1175,6 +1175,12 @@ static inline void hci_remote_features_evt(struct hci_dev *hdev, struct sk_buff
hci_send_cmd(hdev,
HCI_OP_READ_REMOTE_EXT_FEATURES,
sizeof(cp), &cp);
+ } else if (!ev->status && conn->out &&
+ conn->sec_level == BT_SECURITY_HIGH) {
+ struct hci_cp_auth_requested cp;
+ cp.handle = ev->handle;
+ hci_send_cmd(hdev, HCI_OP_AUTH_REQUESTED,
+ sizeof(cp), &cp);
} else {
conn->state = BT_CONNECTED;
hci_proto_connect_cfm(conn, ev->status);
diff --git a/net/bluetooth/hidp/Kconfig b/net/bluetooth/hidp/Kconfig
index 98fdfa1..86a9154 100644
--- a/net/bluetooth/hidp/Kconfig
+++ b/net/bluetooth/hidp/Kconfig
@@ -1,6 +1,6 @@
config BT_HIDP
tristate "HIDP protocol support"
- depends on BT && BT_L2CAP && INPUT
+ depends on BT && BT_L2CAP && INPUT && HID_SUPPORT
select HID
help
HIDP (Human Interface Device Protocol) is a transport layer
diff --git a/net/bluetooth/l2cap.c b/net/bluetooth/l2cap.c
index daa7a98..cd8f6ea 100644
--- a/net/bluetooth/l2cap.c
+++ b/net/bluetooth/l2cap.c
@@ -2421,11 +2421,11 @@ static inline int l2cap_get_conf_opt(void **ptr, int *type, int *olen, unsigned
break;
case 2:
- *val = __le16_to_cpu(*((__le16 *) opt->val));
+ *val = get_unaligned_le16(opt->val);
break;
case 4:
- *val = __le32_to_cpu(*((__le32 *) opt->val));
+ *val = get_unaligned_le32(opt->val);
break;
default:
@@ -2452,11 +2452,11 @@ static void l2cap_add_conf_opt(void **ptr, u8 type, u8 len, unsigned long val)
break;
case 2:
- *((__le16 *) opt->val) = cpu_to_le16(val);
+ put_unaligned_le16(val, opt->val);
break;
case 4:
- *((__le32 *) opt->val) = cpu_to_le32(val);
+ put_unaligned_le32(val, opt->val);
break;
default:
diff --git a/net/bluetooth/rfcomm/core.c b/net/bluetooth/rfcomm/core.c
index 39a5d87..fa642aa 100644
--- a/net/bluetooth/rfcomm/core.c
+++ b/net/bluetooth/rfcomm/core.c
@@ -79,7 +79,10 @@ static void rfcomm_make_uih(struct sk_buff *skb, u8 addr);
static void rfcomm_process_connect(struct rfcomm_session *s);
-static struct rfcomm_session *rfcomm_session_create(bdaddr_t *src, bdaddr_t *dst, int *err);
+static struct rfcomm_session *rfcomm_session_create(bdaddr_t *src,
+ bdaddr_t *dst,
+ u8 sec_level,
+ int *err);
static struct rfcomm_session *rfcomm_session_get(bdaddr_t *src, bdaddr_t *dst);
static void rfcomm_session_del(struct rfcomm_session *s);
@@ -401,7 +404,7 @@ static int __rfcomm_dlc_open(struct rfcomm_dlc *d, bdaddr_t *src, bdaddr_t *dst,
s = rfcomm_session_get(src, dst);
if (!s) {
- s = rfcomm_session_create(src, dst, &err);
+ s = rfcomm_session_create(src, dst, d->sec_level, &err);
if (!s)
return err;
}
@@ -679,7 +682,10 @@ static void rfcomm_session_close(struct rfcomm_session *s, int err)
rfcomm_session_put(s);
}
-static struct rfcomm_session *rfcomm_session_create(bdaddr_t *src, bdaddr_t *dst, int *err)
+static struct rfcomm_session *rfcomm_session_create(bdaddr_t *src,
+ bdaddr_t *dst,
+ u8 sec_level,
+ int *err)
{
struct rfcomm_session *s = NULL;
struct sockaddr_l2 addr;
@@ -704,6 +710,7 @@ static struct rfcomm_session *rfcomm_session_create(bdaddr_t *src, bdaddr_t *dst
sk = sock->sk;
lock_sock(sk);
l2cap_pi(sk)->imtu = l2cap_mtu;
+ l2cap_pi(sk)->sec_level = sec_level;
if (l2cap_ertm)
l2cap_pi(sk)->mode = L2CAP_MODE_ERTM;
release_sock(sk);
diff --git a/net/mac80211/iface.c b/net/mac80211/iface.c
index f9163b1..7aa8559 100644
--- a/net/mac80211/iface.c
+++ b/net/mac80211/iface.c
@@ -391,6 +391,9 @@ static void ieee80211_do_stop(struct ieee80211_sub_if_data *sdata,
u32 hw_reconf_flags = 0;
int i;
+ if (local->scan_sdata == sdata)
+ ieee80211_scan_cancel(local);
+
clear_bit(SDATA_STATE_RUNNING, &sdata->state);
/*
@@ -523,9 +526,6 @@ static void ieee80211_do_stop(struct ieee80211_sub_if_data *sdata,
synchronize_rcu();
skb_queue_purge(&sdata->skb_queue);
- if (local->scan_sdata == sdata)
- ieee80211_scan_cancel(local);
-
/*
* Disable beaconing here for mesh only, AP and IBSS
* are already taken care of.
diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
index c506241..4e78e3f 100644
--- a/net/wireless/nl80211.c
+++ b/net/wireless/nl80211.c
@@ -224,8 +224,8 @@ static int nl80211_prepare_netdev_dump(struct sk_buff *skb,
}
*rdev = cfg80211_get_dev_from_ifindex(sock_net(skb->sk), ifidx);
- if (IS_ERR(dev)) {
- err = PTR_ERR(dev);
+ if (IS_ERR(*rdev)) {
+ err = PTR_ERR(*rdev);
goto out_rtnl;
}
--
John W. Linville Someday the world will need a hero, and you
linville@tuxdriver.com might be all we have. Be ready.
^ permalink raw reply related
* [PATCH] Enhance AF_PACKET implementation to not require high order contiguous memory allocation (v4)
From: nhorman @ 2010-11-10 19:09 UTC (permalink / raw)
To: netdev; +Cc: davem, eric.dumazet, zenczykowski, Neil Horman
In-Reply-To: <1288033566-2091-1-git-send-email-nhorman@tuxdriver.com>
From: Neil Horman <nhorman@tuxdriver.com>
Version 4 of this patch.
Change notes:
1) Removed extra memset. Didn't think kcalloc added a GFP_ZERO the way kzalloc did :)
Summary:
It was shown to me recently that systems under high load were driven very deep
into swap when tcpdump was run. The reason this happened was because the
AF_PACKET protocol has a SET_RINGBUFFER socket option that allows the user space
application to specify how many entries an AF_PACKET socket will have and how
large each entry will be. It seems the default setting for tcpdump is to set
the ring buffer to 32 entries of 64 Kb each, which implies 32 order 5
allocation. Thats difficult under good circumstances, and horrid under memory
pressure.
I thought it would be good to make that a bit more usable. I was going to do a
simple conversion of the ring buffer from contigous pages to iovecs, but
unfortunately, the metadata which AF_PACKET places in these buffers can easily
span a page boundary, and given that these buffers get mapped into user space,
and the data layout doesn't easily allow for a change to padding between frames
to avoid that, a simple iovec change is just going to break user space ABI
consistency.
So I've done this, I've added a three tiered mechanism to the af_packet set_ring
socket option. It attempts to allocate memory in the following order:
1) Using __get_free_pages with GFP_NORETRY set, so as to fail quickly without
digging into swap
2) Using vmalloc
3) Using __get_free_pages with GFP_NORETRY clear, causing us to try as hard as
needed to get the memory
The effect is that we don't disturb the system as much when we're under load,
while still being able to conduct tcpdumps effectively.
Tested successfully by me.
Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
---
net/packet/af_packet.c | 84 ++++++++++++++++++++++++++++++++++++++---------
1 files changed, 68 insertions(+), 16 deletions(-)
diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index 3616f27..a372390 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -163,8 +163,14 @@ struct packet_mreq_max {
static int packet_set_ring(struct sock *sk, struct tpacket_req *req,
int closing, int tx_ring);
+#define PGV_FROM_VMALLOC 1
+struct pgv {
+ char *buffer;
+ unsigned char flags;
+};
+
struct packet_ring_buffer {
- char **pg_vec;
+ struct pgv *pg_vec;
unsigned int head;
unsigned int frames_per_block;
unsigned int frame_size;
@@ -283,7 +289,8 @@ static void *packet_lookup_frame(struct packet_sock *po,
pg_vec_pos = position / rb->frames_per_block;
frame_offset = position % rb->frames_per_block;
- h.raw = rb->pg_vec[pg_vec_pos] + (frame_offset * rb->frame_size);
+ h.raw = rb->pg_vec[pg_vec_pos].buffer +
+ (frame_offset * rb->frame_size);
if (status != __packet_get_status(po, h.raw))
return NULL;
@@ -2322,37 +2329,74 @@ static const struct vm_operations_struct packet_mmap_ops = {
.close = packet_mm_close,
};
-static void free_pg_vec(char **pg_vec, unsigned int order, unsigned int len)
+static void free_pg_vec(struct pgv *pg_vec, unsigned int order,
+ unsigned int len)
{
int i;
for (i = 0; i < len; i++) {
- if (likely(pg_vec[i]))
- free_pages((unsigned long) pg_vec[i], order);
+ if (likely(pg_vec[i].buffer)) {
+ if (pg_vec[i].flags & PGV_FROM_VMALLOC)
+ vfree(pg_vec[i].buffer);
+ else
+ free_pages((unsigned long)pg_vec[i].buffer,
+ order);
+ pg_vec[i].buffer = NULL;
+ }
}
kfree(pg_vec);
}
-static inline char *alloc_one_pg_vec_page(unsigned long order)
+static inline char *alloc_one_pg_vec_page(unsigned long order,
+ unsigned char *flags)
{
- gfp_t gfp_flags = GFP_KERNEL | __GFP_COMP | __GFP_ZERO | __GFP_NOWARN;
+ char *buffer = NULL;
+ gfp_t gfp_flags = GFP_KERNEL | __GFP_COMP |
+ __GFP_ZERO | __GFP_NOWARN | __GFP_NORETRY;
+
+ buffer = (char *) __get_free_pages(gfp_flags, order);
+
+ if (buffer)
+ return buffer;
+
+ /*
+ * __get_free_pages failed, fall back to vmalloc
+ */
+ *flags |= PGV_FROM_VMALLOC;
+ buffer = vmalloc((1 << order) * PAGE_SIZE);
- return (char *) __get_free_pages(gfp_flags, order);
+ if (buffer)
+ return buffer;
+
+ /*
+ * vmalloc failed, lets dig into swap here
+ */
+ *flags = 0;
+ gfp_flags &= ~__GFP_NORETRY;
+ buffer = (char *)__get_free_pages(gfp_flags, order);
+ if (buffer)
+ return buffer;
+
+ /*
+ * complete and utter failure
+ */
+ return NULL;
}
-static char **alloc_pg_vec(struct tpacket_req *req, int order)
+static struct pgv *alloc_pg_vec(struct tpacket_req *req, int order)
{
unsigned int block_nr = req->tp_block_nr;
- char **pg_vec;
+ struct pgv *pg_vec;
int i;
- pg_vec = kzalloc(block_nr * sizeof(char *), GFP_KERNEL);
+ pg_vec = kcalloc(block_nr, sizeof(struct pgv), GFP_KERNEL);
if (unlikely(!pg_vec))
goto out;
for (i = 0; i < block_nr; i++) {
- pg_vec[i] = alloc_one_pg_vec_page(order);
- if (unlikely(!pg_vec[i]))
+ pg_vec[i].buffer = alloc_one_pg_vec_page(order,
+ &pg_vec[i].flags);
+ if (unlikely(!pg_vec[i].buffer))
goto out_free_pgvec;
}
@@ -2361,6 +2405,7 @@ out:
out_free_pgvec:
free_pg_vec(pg_vec, order, block_nr);
+ kfree(pg_vec);
pg_vec = NULL;
goto out;
}
@@ -2368,7 +2413,7 @@ out_free_pgvec:
static int packet_set_ring(struct sock *sk, struct tpacket_req *req,
int closing, int tx_ring)
{
- char **pg_vec = NULL;
+ struct pgv *pg_vec = NULL;
struct packet_sock *po = pkt_sk(sk);
int was_running, order = 0;
struct packet_ring_buffer *rb;
@@ -2530,15 +2575,22 @@ static int packet_mmap(struct file *file, struct socket *sock,
continue;
for (i = 0; i < rb->pg_vec_len; i++) {
- struct page *page = virt_to_page(rb->pg_vec[i]);
+ struct page *page;
+ void *kaddr = rb->pg_vec[i].buffer;
int pg_num;
for (pg_num = 0; pg_num < rb->pg_vec_pages;
- pg_num++, page++) {
+ pg_num++) {
+ if (rb->pg_vec[i].flags & PGV_FROM_VMALLOC)
+ page = vmalloc_to_page(kaddr);
+ else
+ page = virt_to_page(kaddr);
+
err = vm_insert_page(vma, start, page);
if (unlikely(err))
goto out;
start += PAGE_SIZE;
+ kaddr += PAGE_SIZE;
}
}
}
--
1.7.2.3
^ permalink raw reply related
* Re: [net-next PATCH 1/2] qlge: Add firmware info to ethtool get regs.
From: Ron Mercer @ 2010-11-10 19:24 UTC (permalink / raw)
To: David Miller; +Cc: netdev@vger.kernel.org, Jitendra Kalsaria, Ying Ping Lok
In-Reply-To: <20101108.134932.116391208.davem@davemloft.net>
On Mon, Nov 08, 2010 at 01:49:32PM -0800, David Miller wrote:
>
> Nevermind, reverted, can you please more carefully build test
> your changes?
>
> drivers/net/qlge/qlge_mpi.c:90:12: error: static declaration of 'ql_soft_reset_mpi_risc' follows non-static declaration
> drivers/net/qlge/qlge.h:2224:5: note: previous declaration of 'ql_soft_reset_mpi_risc' was here
>
> The line that added the extern declaration to qlge.h seems to be completely
> unrelated to the rest of the patch, as if it's a mis-commit or something
> from another change you were working on.
Dave, I think another patch (correctly) set this function to static
before ours got applied. Corrected patch is on it's way.
Thanks
^ permalink raw reply
* [net-next PATCH 1/2] qlge: Add firmware info to ethtool get regs.
From: Ron Mercer @ 2010-11-10 19:29 UTC (permalink / raw)
To: davem; +Cc: netdev, ron.mercer, jitendra.kalsaria, ying.lok
In-Reply-To: <20101108.134932.116391208.davem@davemloft.net>
By default we add firmware information to ethtool get regs.
Optionally firmware info can instead be sent to log.
Signed-off-by: Jitendra Kalsaria <jitendra.kalsaria@qlogic.com>
Signed-off-by: Ron Mercer <ron.mercer@qlogic.com>
---
drivers/net/qlge/qlge.h | 2 ++
drivers/net/qlge/qlge_dbg.c | 21 ++++++++++++++++++++-
drivers/net/qlge/qlge_ethtool.c | 19 ++++++++++++++++---
drivers/net/qlge/qlge_mpi.c | 2 +-
4 files changed, 39 insertions(+), 5 deletions(-)
diff --git a/drivers/net/qlge/qlge.h b/drivers/net/qlge/qlge.h
index 2282139..b1e8151 100644
--- a/drivers/net/qlge/qlge.h
+++ b/drivers/net/qlge/qlge.h
@@ -2221,6 +2221,7 @@ int ql_write_mpi_reg(struct ql_adapter *qdev, u32 reg, u32 data);
int ql_unpause_mpi_risc(struct ql_adapter *qdev);
int ql_pause_mpi_risc(struct ql_adapter *qdev);
int ql_hard_reset_mpi_risc(struct ql_adapter *qdev);
+int ql_soft_reset_mpi_risc(struct ql_adapter *qdev);
int ql_dump_risc_ram_area(struct ql_adapter *qdev, void *buf,
u32 ram_addr, int word_count);
int ql_core_dump(struct ql_adapter *qdev,
@@ -2236,6 +2237,7 @@ int ql_mb_set_mgmnt_traffic_ctl(struct ql_adapter *qdev, u32 control);
int ql_mb_get_port_cfg(struct ql_adapter *qdev);
int ql_mb_set_port_cfg(struct ql_adapter *qdev);
int ql_wait_fifo_empty(struct ql_adapter *qdev);
+void ql_get_dump(struct ql_adapter *qdev, void *buff);
void ql_gen_reg_dump(struct ql_adapter *qdev,
struct ql_reg_dump *mpi_coredump);
netdev_tx_t ql_lb_send(struct sk_buff *skb, struct net_device *ndev);
diff --git a/drivers/net/qlge/qlge_dbg.c b/drivers/net/qlge/qlge_dbg.c
index 4747492..fca804f 100644
--- a/drivers/net/qlge/qlge_dbg.c
+++ b/drivers/net/qlge/qlge_dbg.c
@@ -1317,9 +1317,28 @@ void ql_gen_reg_dump(struct ql_adapter *qdev,
status = ql_get_ets_regs(qdev, &mpi_coredump->ets[0]);
if (status)
return;
+}
+
+void ql_get_dump(struct ql_adapter *qdev, void *buff)
+{
+ /*
+ * If the dump has already been taken and is stored
+ * in our internal buffer and if force dump is set then
+ * just start the spool to dump it to the log file
+ * and also, take a snapshot of the general regs to
+ * to the user's buffer or else take complete dump
+ * to the user's buffer if force is not set.
+ */
- if (test_bit(QL_FRC_COREDUMP, &qdev->flags))
+ if (!test_bit(QL_FRC_COREDUMP, &qdev->flags)) {
+ if (!ql_core_dump(qdev, buff))
+ ql_soft_reset_mpi_risc(qdev);
+ else
+ netif_err(qdev, drv, qdev->ndev, "coredump failed!\n");
+ } else {
+ ql_gen_reg_dump(qdev, buff);
ql_get_core_dump(qdev);
+ }
}
/* Coredump to messages log file using separate worker thread */
diff --git a/drivers/net/qlge/qlge_ethtool.c b/drivers/net/qlge/qlge_ethtool.c
index 4892d64..8149cc9 100644
--- a/drivers/net/qlge/qlge_ethtool.c
+++ b/drivers/net/qlge/qlge_ethtool.c
@@ -375,7 +375,10 @@ static void ql_get_drvinfo(struct net_device *ndev,
strncpy(drvinfo->bus_info, pci_name(qdev->pdev), 32);
drvinfo->n_stats = 0;
drvinfo->testinfo_len = 0;
- drvinfo->regdump_len = 0;
+ if (!test_bit(QL_FRC_COREDUMP, &qdev->flags))
+ drvinfo->regdump_len = sizeof(struct ql_mpi_coredump);
+ else
+ drvinfo->regdump_len = sizeof(struct ql_reg_dump);
drvinfo->eedump_len = 0;
}
@@ -547,7 +550,12 @@ static void ql_self_test(struct net_device *ndev,
static int ql_get_regs_len(struct net_device *ndev)
{
- return sizeof(struct ql_reg_dump);
+ struct ql_adapter *qdev = netdev_priv(ndev);
+
+ if (!test_bit(QL_FRC_COREDUMP, &qdev->flags))
+ return sizeof(struct ql_mpi_coredump);
+ else
+ return sizeof(struct ql_reg_dump);
}
static void ql_get_regs(struct net_device *ndev,
@@ -555,7 +563,12 @@ static void ql_get_regs(struct net_device *ndev,
{
struct ql_adapter *qdev = netdev_priv(ndev);
- ql_gen_reg_dump(qdev, p);
+ ql_get_dump(qdev, p);
+ qdev->core_is_dumped = 0;
+ if (!test_bit(QL_FRC_COREDUMP, &qdev->flags))
+ regs->len = sizeof(struct ql_mpi_coredump);
+ else
+ regs->len = sizeof(struct ql_reg_dump);
}
static int ql_get_coalesce(struct net_device *dev, struct ethtool_coalesce *c)
diff --git a/drivers/net/qlge/qlge_mpi.c b/drivers/net/qlge/qlge_mpi.c
index 0e7c7c7..100a462 100644
--- a/drivers/net/qlge/qlge_mpi.c
+++ b/drivers/net/qlge/qlge_mpi.c
@@ -87,7 +87,7 @@ exit:
return status;
}
-static int ql_soft_reset_mpi_risc(struct ql_adapter *qdev)
+int ql_soft_reset_mpi_risc(struct ql_adapter *qdev)
{
int status;
status = ql_write_mpi_reg(qdev, 0x00001010, 1);
--
1.6.0.2
^ permalink raw reply related
* [net-next PATCH 2/2] qlge: Version change to v1.00.00.27
From: Ron Mercer @ 2010-11-10 19:29 UTC (permalink / raw)
To: davem; +Cc: netdev, ron.mercer, jitendra.kalsaria, ying.lok
In-Reply-To: <20101108.134932.116391208.davem@davemloft.net>
Signed-off-by: Jitendra Kalsaria <jitendra.kalsaria@qlogic.com>
Signed-off-by: Ron Mercer <ron.mercer@qlogic.com>
---
drivers/net/qlge/qlge.h | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/drivers/net/qlge/qlge.h b/drivers/net/qlge/qlge.h
index b1e8151..bdb8fe8 100644
--- a/drivers/net/qlge/qlge.h
+++ b/drivers/net/qlge/qlge.h
@@ -16,7 +16,7 @@
*/
#define DRV_NAME "qlge"
#define DRV_STRING "QLogic 10 Gigabit PCI-E Ethernet Driver "
-#define DRV_VERSION "v1.00.00.25.00.00-01"
+#define DRV_VERSION "v1.00.00.27.00.00-01"
#define WQ_ADDR_ALIGN 0x3 /* 4 byte alignment */
--
1.6.0.2
^ permalink raw reply related
* Re: [net-next PATCH 1/2] qlge: Add firmware info to ethtool get regs.
From: David Miller @ 2010-11-10 20:07 UTC (permalink / raw)
To: ron.mercer; +Cc: netdev, jitendra.kalsaria, ying.lok
In-Reply-To: <1289417386-28384-1-git-send-email-ron.mercer@qlogic.com>
From: Ron Mercer <ron.mercer@qlogic.com>
Date: Wed, 10 Nov 2010 11:29:45 -0800
> By default we add firmware information to ethtool get regs.
> Optionally firmware info can instead be sent to log.
>
> Signed-off-by: Jitendra Kalsaria <jitendra.kalsaria@qlogic.com>
> Signed-off-by: Ron Mercer <ron.mercer@qlogic.com>
Applied.
^ permalink raw reply
* Re: [net-next PATCH 2/2] qlge: Version change to v1.00.00.27
From: David Miller @ 2010-11-10 20:07 UTC (permalink / raw)
To: ron.mercer; +Cc: netdev, jitendra.kalsaria, ying.lok
In-Reply-To: <1289417386-28384-2-git-send-email-ron.mercer@qlogic.com>
From: Ron Mercer <ron.mercer@qlogic.com>
Date: Wed, 10 Nov 2010 11:29:46 -0800
> Signed-off-by: Jitendra Kalsaria <jitendra.kalsaria@qlogic.com>
> Signed-off-by: Ron Mercer <ron.mercer@qlogic.com>
Applied.
^ permalink raw reply
* Re: [PATCH 2/3 RESEND] net: packet: fix information leak to userland
From: David Miller @ 2010-11-10 20:09 UTC (permalink / raw)
To: segooon; +Cc: kernel-janitors, jpirko, eric.dumazet, netdev, linux-kernel
In-Reply-To: <1289413760-12510-1-git-send-email-segooon@gmail.com>
From: Vasiliy Kulikov <segooon@gmail.com>
Date: Wed, 10 Nov 2010 21:29:18 +0300
> packet_getname_spkt() doesn't initialize all members of sa_data field of
> sockaddr struct if strlen(dev->name) < 13. This structure is then copied
> to userland. It leads to leaking of contents of kernel stack memory.
> We have to fully fill sa_data with strncpy() instead of strlcpy().
>
> The same with packet_getname(): it doesn't initialize sll_pkttype field of
> sockaddr_ll. Set it to zero.
>
> Signed-off-by: Vasiliy Kulikov <segooon@gmail.com>
Applied.
^ permalink raw reply
* Re: [PATCH] net: avoid limits overflow
From: David Miller @ 2010-11-10 20:12 UTC (permalink / raw)
To: eric.dumazet
Cc: holt, akpm, w, linux-kernel, netdev, kuznet, pekkas, jmorris,
yoshfuji, kaber
In-Reply-To: <1289381066.2860.109.camel@edumazet-laptop>
From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Wed, 10 Nov 2010 10:24:26 +0100
> [PATCH] net: avoid limits overflow
>
> Robin Holt tried to boot a 16TB machine and found some limits were
> reached : sysctl_tcp_mem[2], sysctl_udp_mem[2]
>
> We can switch infrastructure to use long "instead" of "int", now
> atomic_long_t primitives are available for free.
>
> Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
> Reported-by: Robin Holt <holt@sgi.com>
> Reviewed-by: Robin Holt <holt@sgi.com>
> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Applied.
^ permalink raw reply
* Re: [PATCH] macvlan: lockless tx path
From: Eric Dumazet @ 2010-11-10 20:33 UTC (permalink / raw)
To: Ben Greear; +Cc: David Miller, Patrick McHardy, netdev
In-Reply-To: <4CDAE713.7020309@candelatech.com>
Le mercredi 10 novembre 2010 à 10:40 -0800, Ben Greear a écrit :
> In my opinion, the kernel and/or driver should deal with this such that
> at worst the user has to deal with 32 v/s 64 bits based on whether the
> kernel is compiled for 32 or 64 bit CPUs. (Let the driver sample at
> intervals needed to never wrap it's counters more than once and update
> software stats of well-defined bit-width, and present those software
> counters to users.
>
How so ? Are you willing to provide patches for all network drivers ?
> In practice, this seems to be the case, at least for the NICs I've used
> (mostly Intel). But, please don't propagate the idea that any width of
> counters is OK to present to user-space: It is completely unfair to
> make app writers have to know the network driver and/or hardware quirks to
> know how often it must sample stats.
>
I am sorry Ben, but /proc/net/dev doesnt publish each counter effective
width. Its unfair, but its like that.
An appplication must be able to cope for wrap arounds, running on a 32
or 64bit kernel. Our duty is to provide 64bit counters for high speed
interfaces where possible.
For a 10Mb adapter, there is no need, since a 32bit counter doesnt wrap
in less than one hour (RFC1902 suggestion)
As I said, many drivers counters are not 32bit or 64bit. I did many
driver get_stats() checks lately...
Why should we cap them to 32bit if they really are 36 or 40 bits ?
> Well, maybe using u32 would have positive benefits on 64-bit kernels then?
>
But we want to handle 40/100Gbps devices, and keep SNMP apps happy.
We really need 64bit for them, and MACVLAN might be used on top of such
devices.
Or are you suggesting using u32 instead of "unsigned long" for
rx_errors/tx_dropped ?
This would indeed save 8 bytes per cpu per macvlan.
^ permalink raw reply
* Re: possible kernel oops from user MSS
From: David Miller @ 2010-11-10 20:41 UTC (permalink / raw)
To: schen; +Cc: netdev
In-Reply-To: <AANLkTimMDMb74-3E9vhFxZ5Dgeuk3HMzPZVjwCj+yFEJ@mail.gmail.com>
From: Steve Chen <schen@mvista.com>
Date: Wed, 10 Nov 2010 07:24:51 -0600
> With commit f5fff5dc8a7a3f395b0525c02ba92c95d42b7390, a user program
> can pass in TCP_MAXSEG of 12 (or TCPOLEN_TSTAMP_ALIGNED), and cause
> kernel oops with division by 0
> in tcp_select_initial_window. One way to prevent it is to change the
> minimum value for TCP_MAXSEG in do_tcp_setsockopt from 8 to some value
> over 12. Two questions.
>
> 1. Is this the right solution?
> 2. If it is, what is a good minimum value?
Thanks Steve, I'll fix this like so:
--------------------
tcp: Increase TCP_MAXSEG socket option minimum.
As noted by Steve Chen, since commit
f5fff5dc8a7a3f395b0525c02ba92c95d42b7390 ("tcp: advertise MSS
requested by user") we can end up with a situation where
tcp_select_initial_window() does a divide by a zero (or
even negative) mss value.
The problem is that sometimes we subtract TCPOLEN_TSTAMP_ALIGNED
from the mss.
Fix this by increasing the minimum from 8 to 8 plus the value
of TCPOLEN_TSTATMP_ALIGNED.
Reported-by: Steve Chen <schen@mvista.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
net/ipv4/tcp.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 245603c..6b0eb4d 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -2246,7 +2246,7 @@ static int do_tcp_setsockopt(struct sock *sk, int level,
/* Values greater than interface MTU won't take effect. However
* at the point when this call is done we typically don't yet
* know which interface is going to be used */
- if (val < 8 || val > MAX_TCP_WINDOW) {
+ if (val < TCPOLEN_TSTAMP_ALIGNED + 8 || val > MAX_TCP_WINDOW) {
err = -EINVAL;
break;
}
--
1.7.3.2
^ permalink raw reply related
* [PATCH net-next-2.6] net: net_families __rcu annotations
From: Eric Dumazet @ 2010-11-10 20:50 UTC (permalink / raw)
To: David Miller; +Cc: netdev
Use modern RCU API / annotations for net_families array.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
---
net/socket.c | 11 ++++++-----
1 file changed, 6 insertions(+), 5 deletions(-)
diff --git a/net/socket.c b/net/socket.c
index 3ca2fd9..c898df7 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -156,7 +156,7 @@ static const struct file_operations socket_file_ops = {
*/
static DEFINE_SPINLOCK(net_family_lock);
-static const struct net_proto_family *net_families[NPROTO] __read_mostly;
+static const struct net_proto_family __rcu *net_families[NPROTO] __read_mostly;
/*
* Statistics counters of the socket lists
@@ -1200,7 +1200,7 @@ int __sock_create(struct net *net, int family, int type, int protocol,
* requested real, full-featured networking support upon configuration.
* Otherwise module support will break!
*/
- if (net_families[family] == NULL)
+ if (rcu_access_pointer(net_families[family]) == NULL)
request_module("net-pf-%d", family);
#endif
@@ -2332,10 +2332,11 @@ int sock_register(const struct net_proto_family *ops)
}
spin_lock(&net_family_lock);
- if (net_families[ops->family])
+ if (rcu_dereference_protected(net_families[ops->family],
+ lockdep_is_held(&net_family_lock)))
err = -EEXIST;
else {
- net_families[ops->family] = ops;
+ rcu_assign_pointer(net_families[ops->family], ops);
err = 0;
}
spin_unlock(&net_family_lock);
@@ -2363,7 +2364,7 @@ void sock_unregister(int family)
BUG_ON(family < 0 || family >= NPROTO);
spin_lock(&net_family_lock);
- net_families[family] = NULL;
+ rcu_assign_pointer(net_families[family], NULL);
spin_unlock(&net_family_lock);
synchronize_rcu();
^ permalink raw reply related
* Re: [PATCH] Fix header size check for GSO case in recvmsg (af_packet)
From: David Miller @ 2010-11-10 20:53 UTC (permalink / raw)
To: mk; +Cc: sri, netdev, linux-kernel
In-Reply-To: <1289253525-7020-1-git-send-email-mk@lab.zgora.pl>
From: Mariusz Kozlowski <mk@lab.zgora.pl>
Date: Mon, 8 Nov 2010 22:58:45 +0100
> Parameter 'len' is size_t type so it will never get negative.
>
> Signed-off-by: Mariusz Kozlowski <mk@lab.zgora.pl>
Applied, thank you!
^ permalink raw reply
* Re: [PATCH] Prevent reading uninitialized memory with socketfilters
From: Ben Hutchings @ 2010-11-10 20:57 UTC (permalink / raw)
To: David Miller; +Cc: penguin-kernel, eric.dumazet, netdev
In-Reply-To: <20101110.103923.59670339.davem@davemloft.net>
On Wed, 2010-11-10 at 10:39 -0800, David Miller wrote:
[...]
> In this patch, I use a bitmap (a single long var) so that only filters
> using mem[] loads/stores pay the price of added security checks.
>
> For other filters, additional cost is a single instruction.
>
> [ Since we access fentry->k a lot now, cache it in a local variable
> and mark filter entry pointer as const. -DaveM ]
[...]
I don't see the justification for combining these changes. One patch,
one fix, right?
Ben.
--
Ben Hutchings, Senior Software Engineer, Solarflare Communications
Not speaking for my employer; that's the marketing department's job.
They asked us to note that Solarflare product names are trademarked.
^ permalink raw reply
* Re: [PATCH] Prevent reading uninitialized memory with socketfilters
From: David Miller @ 2010-11-10 20:59 UTC (permalink / raw)
To: bhutchings; +Cc: penguin-kernel, eric.dumazet, netdev
In-Reply-To: <1289422664.2249.1.camel@achroite.uk.solarflarecom.com>
From: Ben Hutchings <bhutchings@solarflare.com>
Date: Wed, 10 Nov 2010 20:57:44 +0000
> On Wed, 2010-11-10 at 10:39 -0800, David Miller wrote:
> [...]
>> In this patch, I use a bitmap (a single long var) so that only filters
>> using mem[] loads/stores pay the price of added security checks.
>>
>> For other filters, additional cost is a single instruction.
>>
>> [ Since we access fentry->k a lot now, cache it in a local variable
>> and mark filter entry pointer as const. -DaveM ]
> [...]
>
> I don't see the justification for combining these changes. One patch,
> one fix, right?
I'm minimizing the performance impact of the new bitmap checks.
^ permalink raw reply
* Re: [PATCH 0/2] net: Changes in queue allocation and freeing
From: David Miller @ 2010-11-10 21:00 UTC (permalink / raw)
To: therbert; +Cc: eric.dumazet, netdev
In-Reply-To: <AANLkTi=5rxQ6mmnhR7=_dh5u08raaoLiYo=6Gzj9CbL5@mail.gmail.com>
From: Tom Herbert <therbert@google.com>
Date: Wed, 10 Nov 2010 08:27:54 -0800
> Also I noticed that the comment about RX queues refcnts is no longer
> valid. I can respin patch if necessary.
Not necessary, when I apply your patch I'll integrate this comment
removal.
Thanks.
^ permalink raw reply
* Re: [PATCH] macvlan: lockless tx path
From: Ben Hutchings @ 2010-11-10 21:04 UTC (permalink / raw)
To: Eric Dumazet; +Cc: Ben Greear, David Miller, Patrick McHardy, netdev
In-Reply-To: <1289421187.2469.127.camel@edumazet-laptop>
On Wed, 2010-11-10 at 21:33 +0100, Eric Dumazet wrote:
> Le mercredi 10 novembre 2010 à 10:40 -0800, Ben Greear a écrit :
>
> > In my opinion, the kernel and/or driver should deal with this such that
> > at worst the user has to deal with 32 v/s 64 bits based on whether the
> > kernel is compiled for 32 or 64 bit CPUs. (Let the driver sample at
> > intervals needed to never wrap it's counters more than once and update
> > software stats of well-defined bit-width, and present those software
> > counters to users.
> >
>
> How so ? Are you willing to provide patches for all network drivers ?
>
> > In practice, this seems to be the case, at least for the NICs I've used
> > (mostly Intel). But, please don't propagate the idea that any width of
> > counters is OK to present to user-space: It is completely unfair to
> > make app writers have to know the network driver and/or hardware quirks to
> > know how often it must sample stats.
> >
>
> I am sorry Ben, but /proc/net/dev doesnt publish each counter effective
> width. Its unfair, but its like that.
>
> An appplication must be able to cope for wrap arounds, running on a 32
> or 64bit kernel. Our duty is to provide 64bit counters for high speed
> interfaces where possible.
> For a 10Mb adapter, there is no need, since a 32bit counter doesnt wrap
> in less than one hour (RFC1902 suggestion)
>
> As I said, many drivers counters are not 32bit or 64bit. I did many
> driver get_stats() checks lately...
>
> Why should we cap them to 32bit if they really are 36 or 40 bits ?
[...]
Drivers should calculate differences and accumulate them in a 64-bit
counter. (A lot of hardware has read-to-clear counters anyway, in which
case the driver *has* to accumulate the values it reads.)
Ben.
--
Ben Hutchings, Senior Software Engineer, Solarflare Communications
Not speaking for my employer; that's the marketing department's job.
They asked us to note that Solarflare product names are trademarked.
^ permalink raw reply
* Re: [PATCH] macvlan: lockless tx path
From: Eric Dumazet @ 2010-11-10 21:12 UTC (permalink / raw)
To: Ben Hutchings; +Cc: Ben Greear, David Miller, Patrick McHardy, netdev
In-Reply-To: <1289423056.2249.7.camel@achroite.uk.solarflarecom.com>
Le mercredi 10 novembre 2010 à 21:04 +0000, Ben Hutchings a écrit :
> Drivers should calculate differences and accumulate them in a 64-bit
> counter. (A lot of hardware has read-to-clear counters anyway, in which
> case the driver *has* to accumulate the values it reads.)
You are mistaken. These are _hardware_ counters. If they were software,
of course they would be 32 or 64 bit.
And doing the thing you describe in software is racy.
I tried to remove many races, not to add new ones.
Yes, some drivers read one hardware counter using two instructions, and
this is racy.
^ permalink raw reply
* Re: [PATCH] Prevent reading uninitialized memory with socketfilters
From: Ben Hutchings @ 2010-11-10 21:25 UTC (permalink / raw)
To: David Miller; +Cc: penguin-kernel, eric.dumazet, netdev
In-Reply-To: <20101110.125929.245406622.davem@davemloft.net>
On Wed, 2010-11-10 at 12:59 -0800, David Miller wrote:
> From: Ben Hutchings <bhutchings@solarflare.com>
> Date: Wed, 10 Nov 2010 20:57:44 +0000
>
> > On Wed, 2010-11-10 at 10:39 -0800, David Miller wrote:
> > [...]
> >> In this patch, I use a bitmap (a single long var) so that only filters
> >> using mem[] loads/stores pay the price of added security checks.
> >>
> >> For other filters, additional cost is a single instruction.
> >>
> >> [ Since we access fentry->k a lot now, cache it in a local variable
> >> and mark filter entry pointer as const. -DaveM ]
> > [...]
> >
> > I don't see the justification for combining these changes. One patch,
> > one fix, right?
>
> I'm minimizing the performance impact of the new bitmap checks.
This seems like an entirely separate optimisation, since fentry->k was
*already* being used all over the place. (And a smart compiler should
optimise that anyway... though I realise gcc is often not that smart.)
Ben.
--
Ben Hutchings, Senior Software Engineer, Solarflare Communications
Not speaking for my employer; that's the marketing department's job.
They asked us to note that Solarflare product names are trademarked.
^ permalink raw reply
* Re: [PATCH] SUNRPC: Simplify rpc_alloc_iostats by removing pointless local variable
From: Jesper Juhl @ 2010-11-10 21:32 UTC (permalink / raw)
To: linux-kernel-u79uwXL29TY76Z2rM5mHXA
Cc: linux-nfs-u79uwXL29TY76Z2rM5mHXA, netdev-u79uwXL29TY76Z2rM5mHXA,
J. Bruce Fields, Neil Brown, Trond Myklebust, David S. Miller,
Andrew Morton
In-Reply-To: <alpine.LNX.2.00.1011072205370.26247-h2p7t3/P30RzeRGmFJ5qR7ZzlVVXadcDXqFh9Ls21Oc@public.gmane.org>
On Sun, 7 Nov 2010, Jesper Juhl wrote:
> Hi,
>
> We can simplify net/sunrpc/stats.c::rpc_alloc_iostats() a bit by getting
> rid of the unneeded local variable 'new'.
>
>
> Please CC me on replies.
>
>
> Signed-off-by: Jesper Juhl <jj-IYz4IdjRLj0sV2N9l4h3zg@public.gmane.org>
> ---
> stats.c | 4 +---
> 1 file changed, 1 insertion(+), 3 deletions(-)
>
> diff --git
> a/net/sunrpc/stats.c b/net/sunrpc/stats.c
> index f71a731..80df89d 100644
> --- a/net/sunrpc/stats.c
> +++ b/net/sunrpc/stats.c
> @@ -115,9 +115,7 @@ EXPORT_SYMBOL_GPL(svc_seq_show);
> */
> struct rpc_iostats *rpc_alloc_iostats(struct rpc_clnt *clnt)
> {
> - struct rpc_iostats *new;
> - new = kcalloc(clnt->cl_maxproc, sizeof(struct rpc_iostats), GFP_KERNEL);
> - return new;
> + return kcalloc(clnt->cl_maxproc, sizeof(struct rpc_iostats), GFP_KERNEL);
> }
> EXPORT_SYMBOL_GPL(rpc_alloc_iostats);
>
>
>
Ok, no response to this for a couple of days.
Is there some problem or did it just get missed?
Could someone merge this and push it up-stream, please, if there are no
problems with it...
--
Jesper Juhl <jj-IYz4IdjRLj0sV2N9l4h3zg@public.gmane.org> http://www.chaosbits.net/
Don't top-post http://www.catb.org/~esr/jargon/html/T/top-post.html
Plain text mails only, please.
--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply
* Re: [PATCH] macvlan: lockless tx path
From: Ben Greear @ 2010-11-10 21:35 UTC (permalink / raw)
To: Eric Dumazet; +Cc: David Miller, Patrick McHardy, netdev
In-Reply-To: <1289421187.2469.127.camel@edumazet-laptop>
On 11/10/2010 12:33 PM, Eric Dumazet wrote:
> Le mercredi 10 novembre 2010 à 10:40 -0800, Ben Greear a écrit :
>
>> In my opinion, the kernel and/or driver should deal with this such that
>> at worst the user has to deal with 32 v/s 64 bits based on whether the
>> kernel is compiled for 32 or 64 bit CPUs. (Let the driver sample at
>> intervals needed to never wrap it's counters more than once and update
>> software stats of well-defined bit-width, and present those software
>> counters to users.
>>
>
> How so ? Are you willing to provide patches for all network drivers ?
I'm willing to attempt to fix something that I use and can test.
Either way, I think it's legitimate to document at least the desired
behaviour so that driver writers know what to aim for.
>> In practice, this seems to be the case, at least for the NICs I've used
>> (mostly Intel). But, please don't propagate the idea that any width of
>> counters is OK to present to user-space: It is completely unfair to
>> make app writers have to know the network driver and/or hardware quirks to
>> know how often it must sample stats.
>>
>
> I am sorry Ben, but /proc/net/dev doesnt publish each counter effective
> width. Its unfair, but its like that.
>
> An appplication must be able to cope for wrap arounds, running on a 32
> or 64bit kernel. Our duty is to provide 64bit counters for high speed
> interfaces where possible.
> For a 10Mb adapter, there is no need, since a 32bit counter doesnt wrap
> in less than one hour (RFC1902 suggestion)
So an application that must deal with wraps must poll at the minimal
time interval for wrapping 32-bit counters at whatever speed, or it
must pay attention to the driver to somehow know that this magic driver
can *really* do 64-bit stats properly?
Please note that just because a counter is less than the previous read,
that doesn't by itself tell us if it wrapped once or twice. And, if we
don't know at which number of bits it wraps, then we don't know how many
to add even if we are certain it wrapped only once.
In general, I want to treat eth0 the same as eth5, and not worry that one
is 10/100 realtek and the other a 10G Intel.
If netlink reports stats64, then those should only wrap at 64 bits,
and if it reports stats32, then wrap at 32-bits.
> As I said, many drivers counters are not 32bit or 64bit. I did many
> driver get_stats() checks lately...
>
> Why should we cap them to 32bit if they really are 36 or 40 bits ?
>
>
>> Well, maybe using u32 would have positive benefits on 64-bit kernels then?
>>
>
> But we want to handle 40/100Gbps devices, and keep SNMP apps happy.
>
> We really need 64bit for them, and MACVLAN might be used on top of such
> devices.
>
> Or are you suggesting using u32 instead of "unsigned long" for
> rx_errors/tx_dropped ?
>
> This would indeed save 8 bytes per cpu per macvlan.
Yes, that was what I was trying to suggest. I'm all for 64-bit numbers
in anything that can wrap anytime soon, and anywhere you think 32-bits
is enough, just use u32 so we don't have to worry about the number of
bits in 'unsigned long' on different platforms.
Thanks,
Ben
--
Ben Greear <greearb@candelatech.com>
Candela Technologies Inc http://www.candelatech.com
^ permalink raw reply
* Re: [PATCH] macvlan: lockless tx path
From: Ben Hutchings @ 2010-11-10 21:53 UTC (permalink / raw)
To: Eric Dumazet; +Cc: Ben Greear, David Miller, Patrick McHardy, netdev
In-Reply-To: <1289423555.17691.8.camel@edumazet-laptop>
On Wed, 2010-11-10 at 22:12 +0100, Eric Dumazet wrote:
> Le mercredi 10 novembre 2010 à 21:04 +0000, Ben Hutchings a écrit :
>
> > Drivers should calculate differences and accumulate them in a 64-bit
> > counter. (A lot of hardware has read-to-clear counters anyway, in which
> > case the driver *has* to accumulate the values it reads.)
>
> You are mistaken. These are _hardware_ counters. If they were software,
> of course they would be 32 or 64 bit.
Of course I understood that.
> And doing the thing you describe in software is racy.
> I tried to remove many races, not to add new ones.
If you do it in ndo_get_stats{,64} and don't use your own lock, yes.
> Yes, some drivers read one hardware counter using two instructions, and
> this is racy.
Most hardware which supports MMIO to multi-word counters has some kind
of latching scheme where you can read the words/registers in order and
get consistent values (within a single counter; consistency between
counters is another matter). Obviously you have to use a lock to
serialise stats updates in this case - whether or not you maintain wider
software counters.
Oh, another problem with using register values directly is that
statistics are likely to be reset whenever the device is reconfigured in
the way that requires a hardware reset.
Ben.
--
Ben Hutchings, Senior Software Engineer, Solarflare Communications
Not speaking for my employer; that's the marketing department's job.
They asked us to note that Solarflare product names are trademarked.
^ permalink raw reply
* Re: [PATCH] Fix CAN info leak/minor heap overflow
From: Oliver Hartkopp @ 2010-11-10 22:10 UTC (permalink / raw)
To: David Miller; +Cc: urs, netdev, drosenberg, security, torvalds
In-Reply-To: <20101110.095141.226780406.davem@davemloft.net>
On 10.11.2010 18:51, David Miller wrote:
> From: Oliver Hartkopp <socketcan@hartkopp.net>
> Date: Wed, 10 Nov 2010 07:52:27 +0100
>
>> IMHO the patch improves the historic situation and fixes the useless leakage
>> of kernel addresses. Please consider to apply that procfs changes.
>
> I'm only fine with fixing the kernel pointer fields in some way.
>
> But moving forward any other change to the procfs file is simply
> a waste of time.
>
> You should create sysfs files and add logic to your tools to look
> for them and use them if they exist.
>
> Your forward path _SHOULD NOT_ be continuing this procfs versioning
> madness. Use something sane and do the work to make userland start
> to be ready for this transition.
Hm, summarizing the given restrictions and taking into account that just
setting the pointer fields to '0' is said to be annoying, the only thing that
can be fixed is the minor heap overflow caused by the char array. I'll send a
patch for that.
As you don't want to change the layout even if there's no tool relying on the
entries i wanted to modify, i'll just stop my attempts to improve it.
Regards,
Oliver
^ permalink raw reply
* can-bcm: fix minor heap overflow
From: Oliver Hartkopp @ 2010-11-10 22:10 UTC (permalink / raw)
To: David Miller
Cc: Linux Netdev List, Dan Rosenberg, Linus Torvalds, Urs Thuermann,
security
On 64-bit platforms the ASCII representation of a pointer may be up to 17
bytes long. This patch increases the length of the buffer accordingly.
http://marc.info/?l=linux-netdev&m=128872251418192&w=2
Reported-by: Dan Rosenberg <drosenberg@vsecurity.com>
Signed-off-by: Oliver Hartkopp <socketcan@hartkopp.net>
CC: Linus Torvalds <torvalds@linux-foundation.org>
---
diff --git a/net/can/bcm.c b/net/can/bcm.c
index 08ffe9e..6faa825 100644
--- a/net/can/bcm.c
+++ b/net/can/bcm.c
@@ -125,7 +125,7 @@ struct bcm_sock {
struct list_head tx_ops;
unsigned long dropped_usr_msgs;
struct proc_dir_entry *bcm_proc_read;
- char procname [9]; /* pointer printed in ASCII with \0 */
+ char procname [20]; /* pointer printed in ASCII with \0 */
};
static inline struct bcm_sock *bcm_sk(const struct sock *sk)
^ permalink raw reply related
* Re: [PATCH] macvlan: lockless tx path
From: Eric Dumazet @ 2010-11-10 22:21 UTC (permalink / raw)
To: Ben Greear; +Cc: David Miller, Patrick McHardy, netdev
In-Reply-To: <4CDB1021.507@candelatech.com>
Le mercredi 10 novembre 2010 à 13:35 -0800, Ben Greear a écrit :
> So an application that must deal with wraps must poll at the minimal
> time interval for wrapping 32-bit counters at whatever speed, or it
> must pay attention to the driver to somehow know that this magic driver
> can *really* do 64-bit stats properly?
>
Are you aware that you speak of something that is not specified at all
in linux ?
Frequency of polling is not part of any RFC. This usually is tunable in
the _application_. Some people sample stats every 5 minutes, some sample
every second, and hit the "xxx driver updates its stats every two
seconds, this sucks"
I wrote SNMP apps based on /proc/net/dev and all just work, with any
versions, any driver. Of course, some of them broke 6 years ago because
they were 32bit legacy application, running on a 64bit kernel. I never
asked David to change /proc/net/dev to cap counters to 32bit.
When 128bit cpu come, some userland changes are needed to parse 128bit
numbers.
In anycase, apps dont have to know a particular driver provides 64bit or
32bit counter. Only choice for them is to automatically detect the
wraparound, because they fetch a STRING, not a Counter32 or Counter64
This works for all drivers, legacy, new, Intel or whatever. If a driver
changes from 32 to 64, nothing special happens in /proc/net/dev.
RRD for example handles this just fine.
> Please note that just because a counter is less than the previous read,
> that doesn't by itself tell us if it wrapped once or twice. And, if we
> don't know at which number of bits it wraps, then we don't know how many
> to add even if we are certain it wrapped only once.
>
I repeat : Nothing in /proc/net/dev can tell you when a counter will
wrap (the counter width).
You also need to use the correct polling frequency, depending on max
speed. It was already the case with 32bit counters, 64bit ones only gave
some extra range.
> In general, I want to treat eth0 the same as eth5, and not worry that one
> is 10/100 realtek and the other a 10G Intel.
>
> If netlink reports stats64, then those should only wrap at 64 bits,
> and if it reports stats32, then wrap at 32-bits.
>
I believe you are mistaken. We provide stats64 for all drivers, even
32bit legacy ones. rtnetlink has no way to report counter widths,
because nobody cared.
^ 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