* [PATCH RESEND net 2/9] net/mac89x0: Remove dead or unreachable code
From: Finn Thain @ 2017-10-03 1:07 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, linux-kernel
In-Reply-To: <cover.1506992619.git.fthain@telegraphics.com.au>
Signed-off-by: Finn Thain <fthain@telegraphics.com.au>
---
drivers/net/ethernet/cirrus/mac89x0.c | 36 -----------------------------------
1 file changed, 36 deletions(-)
diff --git a/drivers/net/ethernet/cirrus/mac89x0.c b/drivers/net/ethernet/cirrus/mac89x0.c
index f910f0f386d6..8fc43c865621 100644
--- a/drivers/net/ethernet/cirrus/mac89x0.c
+++ b/drivers/net/ethernet/cirrus/mac89x0.c
@@ -119,10 +119,6 @@ struct net_local {
};
/* Index to functions, as function prototypes. */
-
-#if 0
-extern void reset_chip(struct net_device *dev);
-#endif
static int net_open(struct net_device *dev);
static int net_send_packet(struct sk_buff *skb, struct net_device *dev);
static irqreturn_t net_interrupt(int irq, void *dev_id);
@@ -132,10 +128,6 @@ static int net_close(struct net_device *dev);
static struct net_device_stats *net_get_stats(struct net_device *dev);
static int set_mac_address(struct net_device *dev, void *addr);
-
-/* Example routines you must write ;->. */
-#define tx_done(dev) 1
-
/* For reading/writing registers ISA-style */
static inline int
readreg_io(struct net_device *dev, int portno)
@@ -179,7 +171,6 @@ static const struct net_device_ops mac89x0_netdev_ops = {
struct net_device * __init mac89x0_probe(int unit)
{
struct net_device *dev;
- static int once_is_enough;
struct net_local *lp;
static unsigned version_printed;
int i, slot;
@@ -200,10 +191,6 @@ struct net_device * __init mac89x0_probe(int unit)
netdev_boot_setup_check(dev);
}
- if (once_is_enough)
- goto out;
- once_is_enough = 1;
-
/* We might have to parameterize this later */
slot = 0xE;
/* Get out now if there's a real NuBus card in slot E */
@@ -295,24 +282,6 @@ struct net_device * __init mac89x0_probe(int unit)
return ERR_PTR(err);
}
-#if 0
-/* This is useful for something, but I don't know what yet. */
-void __init reset_chip(struct net_device *dev)
-{
- int reset_start_time;
-
- writereg(dev, PP_SelfCTL, readreg(dev, PP_SelfCTL) | POWER_ON_RESET);
-
- /* wait 30 ms */
- msleep_interruptible(30);
-
- /* Wait until the chip is reset */
- reset_start_time = jiffies;
- while( (readreg(dev, PP_SelfST) & INIT_DONE) == 0 && jiffies - reset_start_time < 2)
- ;
-}
-#endif
-
/* Open/initialize the board. This is called (in the current kernel)
sometime after booting when the 'ifconfig' program is run.
@@ -414,11 +383,6 @@ static irqreturn_t net_interrupt(int irq, void *dev_id)
struct net_local *lp;
int ioaddr, status;
- if (dev == NULL) {
- printk ("net_interrupt(): irq %d for unknown device.\n", irq);
- return IRQ_NONE;
- }
-
ioaddr = dev->base_addr;
lp = netdev_priv(dev);
--
2.13.5
^ permalink raw reply related
* [PATCH RESEND net 4/9] net/mac89x0: Replace custom debug logging with netif_* calls
From: Finn Thain @ 2017-10-03 1:07 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, linux-kernel
In-Reply-To: <cover.1506992619.git.fthain@telegraphics.com.au>
Signed-off-by: Finn Thain <fthain@telegraphics.com.au>
---
drivers/net/ethernet/cirrus/mac89x0.c | 47 ++++++++++++-----------------------
1 file changed, 16 insertions(+), 31 deletions(-)
diff --git a/drivers/net/ethernet/cirrus/mac89x0.c b/drivers/net/ethernet/cirrus/mac89x0.c
index 4393d9b89f33..278df230c5f6 100644
--- a/drivers/net/ethernet/cirrus/mac89x0.c
+++ b/drivers/net/ethernet/cirrus/mac89x0.c
@@ -61,18 +61,6 @@
static const char version[] =
"cs89x0.c:v1.02 11/26/96 Russell Nelson <nelson@crynwr.com>\n";
-/* ======================= configure the driver here ======================= */
-
-/* use 0 for production, 1 for verification, >2 for debug */
-#ifndef NET_DEBUG
-#define NET_DEBUG 0
-#endif
-
-/* ======================= end of configuration ======================= */
-
-
-/* Always include 'config.h' first in case the user wants to turn on
- or override something. */
#include <linux/module.h>
/*
@@ -107,10 +95,13 @@ static const char version[] =
#include "cs89x0.h"
-static unsigned int net_debug = NET_DEBUG;
+static int debug = -1;
+module_param(debug, int, 0);
+MODULE_PARM_DESC(debug, "debug message level");
/* Information that need to be kept for each board. */
struct net_local {
+ int msg_enable;
int chip_type; /* one of: CS8900, CS8920, CS8920M */
char chip_revision; /* revision letter of the chip ('A'...) */
int send_cmd; /* the propercommand used to send a packet. */
@@ -174,7 +165,6 @@ struct net_device * __init mac89x0_probe(int unit)
{
struct net_device *dev;
struct net_local *lp;
- static unsigned version_printed;
int i, slot;
unsigned rev_type = 0;
unsigned long ioaddr;
@@ -220,6 +210,8 @@ struct net_device * __init mac89x0_probe(int unit)
/* Initialize the net_device structure. */
lp = netdev_priv(dev);
+ lp->msg_enable = netif_msg_init(debug, 0);
+
/* Fill in the 'dev' fields. */
dev->base_addr = ioaddr;
dev->mem_start = (unsigned long)
@@ -242,8 +234,7 @@ struct net_device * __init mac89x0_probe(int unit)
if (lp->chip_type != CS8900 && lp->chip_revision >= 'C')
lp->send_cmd = TX_NOW;
- if (net_debug && version_printed++ == 0)
- printk(version);
+ netif_dbg(lp, drv, dev, "%s", version);
pr_info("CS89%c0%s rev %c found at %#8lx\n",
lp->chip_type == CS8900 ? '0' : '2',
@@ -340,11 +331,9 @@ net_send_packet(struct sk_buff *skb, struct net_device *dev)
struct net_local *lp = netdev_priv(dev);
unsigned long flags;
- if (net_debug > 3)
- printk("%s: sent %d byte packet of type %x\n",
- dev->name, skb->len,
- (skb->data[ETH_ALEN+ETH_ALEN] << 8)
- | skb->data[ETH_ALEN+ETH_ALEN+1]);
+ netif_dbg(lp, tx_queued, dev, "sent %d byte packet of type %x\n",
+ skb->len, skb->data[ETH_ALEN + ETH_ALEN] << 8 |
+ skb->data[ETH_ALEN + ETH_ALEN + 1]);
/* keep the upload from being interrupted, since we
ask the chip to start transmitting before the
@@ -393,7 +382,7 @@ static irqreturn_t net_interrupt(int irq, void *dev_id)
faster than you can read them off, you're screwed. Hasta la
vista, baby! */
while ((status = swab16(nubus_readw(dev->base_addr + ISQ_PORT)))) {
- if (net_debug > 4)printk("%s: event=%04x\n", dev->name, status);
+ netif_dbg(lp, intr, dev, "status=%04x\n", status);
switch(status & ISQ_EVENT_MASK) {
case ISQ_RECEIVER_EVENT:
/* Got a packet(s). */
@@ -423,7 +412,7 @@ static irqreturn_t net_interrupt(int irq, void *dev_id)
netif_wake_queue(dev);
}
if (status & TX_UNDERRUN) {
- if (net_debug > 0) printk("%s: transmit underrun\n", dev->name);
+ netif_dbg(lp, tx_err, dev, "transmit underrun\n");
lp->send_underrun++;
if (lp->send_underrun == 3) lp->send_cmd = TX_AFTER_381;
else if (lp->send_underrun == 6) lp->send_cmd = TX_AFTER_ALL;
@@ -444,6 +433,7 @@ static irqreturn_t net_interrupt(int irq, void *dev_id)
static void
net_rx(struct net_device *dev)
{
+ struct net_local *lp = netdev_priv(dev);
struct sk_buff *skb;
int status, length;
@@ -475,10 +465,9 @@ net_rx(struct net_device *dev)
skb_copy_to_linear_data(skb, (void *)(dev->mem_start + PP_RxFrame),
length);
- if (net_debug > 3)printk("%s: received %d byte packet of type %x\n",
- dev->name, length,
- (skb->data[ETH_ALEN+ETH_ALEN] << 8)
- | skb->data[ETH_ALEN+ETH_ALEN+1]);
+ netif_dbg(lp, rx_status, dev, "received %d byte packet of type %x\n",
+ length, skb->data[ETH_ALEN + ETH_ALEN] << 8 |
+ skb->data[ETH_ALEN + ETH_ALEN + 1]);
skb->protocol=eth_type_trans(skb,dev);
netif_rx(skb);
@@ -566,16 +555,12 @@ static int set_mac_address(struct net_device *dev, void *addr)
#ifdef MODULE
static struct net_device *dev_cs89x0;
-static int debug;
-module_param(debug, int, 0);
-MODULE_PARM_DESC(debug, "CS89[02]0 debug level (0-5)");
MODULE_LICENSE("GPL");
int __init
init_module(void)
{
- net_debug = debug;
dev_cs89x0 = mac89x0_probe(-1);
if (IS_ERR(dev_cs89x0)) {
pr_warn("No card found\n");
--
2.13.5
^ permalink raw reply related
* [PATCH RESEND net 3/9] net/mac89x0: Fix and modernize log messages
From: Finn Thain @ 2017-10-03 1:07 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, linux-kernel
In-Reply-To: <cover.1506992619.git.fthain@telegraphics.com.au>
Fix misplaced newlines in conditional log messages.
Add missing printk severity levels.
Log the MAC address after the interface gets a meaningful name.
Drop deprecated "out of memory" message as per checkpatch advice.
Signed-off-by: Finn Thain <fthain@telegraphics.com.au>
---
drivers/net/ethernet/cirrus/mac89x0.c | 26 ++++++++++++--------------
1 file changed, 12 insertions(+), 14 deletions(-)
diff --git a/drivers/net/ethernet/cirrus/mac89x0.c b/drivers/net/ethernet/cirrus/mac89x0.c
index 8fc43c865621..4393d9b89f33 100644
--- a/drivers/net/ethernet/cirrus/mac89x0.c
+++ b/drivers/net/ethernet/cirrus/mac89x0.c
@@ -56,6 +56,8 @@
local_irq_{dis,en}able()
*/
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
static const char version[] =
"cs89x0.c:v1.02 11/26/96 Russell Nelson <nelson@crynwr.com>\n";
@@ -243,16 +245,14 @@ struct net_device * __init mac89x0_probe(int unit)
if (net_debug && version_printed++ == 0)
printk(version);
- printk(KERN_INFO "%s: cs89%c0%s rev %c found at %#8lx",
- dev->name,
- lp->chip_type==CS8900?'0':'2',
- lp->chip_type==CS8920M?"M":"",
- lp->chip_revision,
- dev->base_addr);
+ pr_info("CS89%c0%s rev %c found at %#8lx\n",
+ lp->chip_type == CS8900 ? '0' : '2',
+ lp->chip_type == CS8920M ? "M" : "",
+ lp->chip_revision, dev->base_addr);
/* Try to read the MAC address */
if ((readreg(dev, PP_SelfST) & (EEPROM_PRESENT | EEPROM_OK)) == 0) {
- printk("\nmac89x0: No EEPROM, giving up now.\n");
+ pr_info("No EEPROM, giving up now\n");
goto out1;
} else {
for (i = 0; i < ETH_ALEN; i += 2) {
@@ -265,15 +265,14 @@ struct net_device * __init mac89x0_probe(int unit)
dev->irq = SLOT2IRQ(slot);
- /* print the IRQ and ethernet address. */
-
- printk(" IRQ %d ADDR %pM\n", dev->irq, dev->dev_addr);
-
dev->netdev_ops = &mac89x0_netdev_ops;
err = register_netdev(dev);
if (err)
goto out1;
+
+ netdev_info(dev, "MAC %pM, IRQ %d\n", dev->dev_addr, dev->irq);
+
return NULL;
out1:
nubus_writew(0, dev->base_addr + ADD_PORT);
@@ -468,7 +467,6 @@ net_rx(struct net_device *dev)
/* Malloc up new buffer. */
skb = alloc_skb(length, GFP_ATOMIC);
if (skb == NULL) {
- printk("%s: Memory squeeze, dropping packet.\n", dev->name);
dev->stats.rx_dropped++;
return;
}
@@ -556,7 +554,7 @@ static int set_mac_address(struct net_device *dev, void *addr)
return -EADDRNOTAVAIL;
memcpy(dev->dev_addr, saddr->sa_data, ETH_ALEN);
- printk("%s: Setting MAC address to %pM\n", dev->name, dev->dev_addr);
+ netdev_info(dev, "Setting MAC address to %pM\n", dev->dev_addr);
/* set the Ethernet address */
for (i=0; i < ETH_ALEN/2; i++)
@@ -580,7 +578,7 @@ init_module(void)
net_debug = debug;
dev_cs89x0 = mac89x0_probe(-1);
if (IS_ERR(dev_cs89x0)) {
- printk(KERN_WARNING "mac89x0.c: No card found\n");
+ pr_warn("No card found\n");
return PTR_ERR(dev_cs89x0);
}
return 0;
--
2.13.5
^ permalink raw reply related
* [PATCH RESEND net 6/9] net/sonic: Cleanup and modernize log messages
From: Finn Thain @ 2017-10-03 1:07 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, linux-kernel, Thomas Bogendoerfer, Chris Zankel
In-Reply-To: <cover.1506992619.git.fthain@telegraphics.com.au>
Add missing printk severity levels.
Don't print the valid silicon revision number (it's in the source).
Avoid KERN_CONT usage as per advice from checkpatch.
Avoid ifdef around printk calls.
Cc: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
Cc: Chris Zankel <chris@zankel.net>
Signed-off-by: Finn Thain <fthain@telegraphics.com.au>
---
drivers/net/ethernet/natsemi/jazzsonic.c | 9 +++------
drivers/net/ethernet/natsemi/macsonic.c | 24 ++++++++++--------------
drivers/net/ethernet/natsemi/xtsonic.c | 11 ++++-------
3 files changed, 17 insertions(+), 27 deletions(-)
diff --git a/drivers/net/ethernet/natsemi/jazzsonic.c b/drivers/net/ethernet/natsemi/jazzsonic.c
index a6caeb567c0d..6ec25c27cb73 100644
--- a/drivers/net/ethernet/natsemi/jazzsonic.c
+++ b/drivers/net/ethernet/natsemi/jazzsonic.c
@@ -132,17 +132,14 @@ static int sonic_probe1(struct net_device *dev)
* the expected location.
*/
silicon_revision = SONIC_READ(SONIC_SR);
- if (sonic_debug > 1)
- printk("SONIC Silicon Revision = 0x%04x\n",silicon_revision);
-
i = 0;
while (known_revisions[i] != 0xffff &&
known_revisions[i] != silicon_revision)
i++;
if (known_revisions[i] == 0xffff) {
- printk("SONIC ethernet controller not found (0x%4x)\n",
- silicon_revision);
+ pr_info("SONIC ethernet controller not found (0x%4x)\n",
+ silicon_revision);
goto out;
}
@@ -248,7 +245,7 @@ static int jazz_sonic_probe(struct platform_device *pdev)
if (err)
goto out1;
- printk("%s: MAC %pM IRQ %d\n", dev->name, dev->dev_addr, dev->irq);
+ netdev_info(dev, "MAC %pM, IRQ %d\n", dev->dev_addr, dev->irq);
return 0;
diff --git a/drivers/net/ethernet/natsemi/macsonic.c b/drivers/net/ethernet/natsemi/macsonic.c
index 3ca6ae7caf55..795d71522617 100644
--- a/drivers/net/ethernet/natsemi/macsonic.c
+++ b/drivers/net/ethernet/natsemi/macsonic.c
@@ -315,8 +315,6 @@ static int mac_onboard_sonic_probe(struct net_device *dev)
if (!MACH_IS_MAC)
return -ENODEV;
- printk(KERN_INFO "Checking for internal Macintosh ethernet (SONIC).. ");
-
/* Bogus probing, on the models which may or may not have
Ethernet (BTW, the Ethernet *is* always at the same
address, and nothing else lives there, at least if Apple's
@@ -329,14 +327,12 @@ static int mac_onboard_sonic_probe(struct net_device *dev)
card_present = hwreg_present((void*)ONBOARD_SONIC_REGISTERS);
if (!card_present) {
- printk("none.\n");
+ pr_info("No onboard or comm-slot SONIC was detected\n");
return -ENODEV;
}
commslot = 1;
}
- printk("yes\n");
-
/* Danger! My arms are flailing wildly! You *must* set lp->reg_offset
* and dev->base_addr before using SONIC_READ() or SONIC_WRITE() */
dev->base_addr = ONBOARD_SONIC_REGISTERS;
@@ -385,10 +381,10 @@ static int mac_onboard_sonic_probe(struct net_device *dev)
"%s: revision 0x%04x, using %d bit DMA and register offset %d\n",
dev_name(lp->device), sr, lp->dma_bitmode?32:16, lp->reg_offset);
-#if 0 /* This is sometimes useful to find out how MacOS configured the card. */
- printk(KERN_INFO "%s: DCR: 0x%04x, DCR2: 0x%04x\n", dev_name(lp->device),
- SONIC_READ(SONIC_DCR) & 0xffff, SONIC_READ(SONIC_DCR2) & 0xffff);
-#endif
+ /* This is sometimes useful to find out how MacOS configured the card */
+ pr_debug("%s: DCR=0x%04x, DCR2=0x%04x\n", __func__,
+ SONIC_READ(SONIC_DCR) & 0xffff,
+ SONIC_READ(SONIC_DCR2) & 0xffff);
/* Software reset, then initialize control registers. */
SONIC_WRITE(SONIC_CMD, SONIC_CR_RST);
@@ -540,10 +536,10 @@ static int mac_nubus_sonic_probe(struct net_device *dev)
printk(KERN_INFO "%s: revision 0x%04x, using %d bit DMA and register offset %d\n",
dev_name(lp->device), SONIC_READ(SONIC_SR), dma_bitmode?32:16, reg_offset);
-#if 0 /* This is sometimes useful to find out how MacOS configured the card. */
- printk(KERN_INFO "%s: DCR: 0x%04x, DCR2: 0x%04x\n", dev_name(lp->device),
- SONIC_READ(SONIC_DCR) & 0xffff, SONIC_READ(SONIC_DCR2) & 0xffff);
-#endif
+ /* This is sometimes useful to find out how MacOS configured the card */
+ pr_debug("%s: DCR=0x%04x, DCR2=0x%04x\n", __func__,
+ SONIC_READ(SONIC_DCR) & 0xffff,
+ SONIC_READ(SONIC_DCR2) & 0xffff);
/* Software reset, then initialize control registers. */
SONIC_WRITE(SONIC_CMD, SONIC_CR_RST);
@@ -594,7 +590,7 @@ static int mac_sonic_probe(struct platform_device *pdev)
if (err)
goto out;
- printk("%s: MAC %pM IRQ %d\n", dev->name, dev->dev_addr, dev->irq);
+ netdev_info(dev, "MAC %pM, IRQ %d\n", dev->dev_addr, dev->irq);
return 0;
diff --git a/drivers/net/ethernet/natsemi/xtsonic.c b/drivers/net/ethernet/natsemi/xtsonic.c
index 9ee0f69a83c0..bed89e8105e9 100644
--- a/drivers/net/ethernet/natsemi/xtsonic.c
+++ b/drivers/net/ethernet/natsemi/xtsonic.c
@@ -145,17 +145,14 @@ static int __init sonic_probe1(struct net_device *dev)
* the expected location.
*/
silicon_revision = SONIC_READ(SONIC_SR);
- if (sonic_debug > 1)
- printk("SONIC Silicon Revision = 0x%04x\n",silicon_revision);
-
i = 0;
while ((known_revisions[i] != 0xffff) &&
(known_revisions[i] != silicon_revision))
i++;
if (known_revisions[i] == 0xffff) {
- printk("SONIC ethernet controller not found (0x%4x)\n",
- silicon_revision);
+ pr_info("SONIC ethernet controller not found (0x%4x)\n",
+ silicon_revision);
return -ENODEV;
}
@@ -275,8 +272,8 @@ int xtsonic_probe(struct platform_device *pdev)
if ((err = register_netdev(dev)))
goto out1;
- printk("%s: SONIC ethernet @%08lx, MAC %pM, IRQ %d\n", dev->name,
- dev->base_addr, dev->dev_addr, dev->irq);
+ netdev_info(dev, "SONIC ethernet @%08lx, MAC %pM, IRQ %d\n",
+ dev->base_addr, dev->dev_addr, dev->irq);
return 0;
--
2.13.5
^ permalink raw reply related
* [PATCH RESEND net 5/9] net/macmace: Fix and cleanup log messages
From: Finn Thain @ 2017-10-03 1:07 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, linux-kernel
In-Reply-To: <cover.1506992619.git.fthain@telegraphics.com.au>
Log the MAC address after the interface gets a meaningful name.
Drop redundant debug messages for FIFO events recorded in the
interface statistics (consistent with mace.c).
Signed-off-by: Finn Thain <fthain@telegraphics.com.au>
---
drivers/net/ethernet/apple/macmace.c | 10 +++-------
1 file changed, 3 insertions(+), 7 deletions(-)
diff --git a/drivers/net/ethernet/apple/macmace.c b/drivers/net/ethernet/apple/macmace.c
index f17a160dbff2..3ff53c3ef732 100644
--- a/drivers/net/ethernet/apple/macmace.c
+++ b/drivers/net/ethernet/apple/macmace.c
@@ -247,13 +247,12 @@ static int mace_probe(struct platform_device *pdev)
dev->netdev_ops = &mace_netdev_ops;
dev->watchdog_timeo = TX_TIMEOUT;
- printk(KERN_INFO "%s: 68K MACE, hardware address %pM\n",
- dev->name, dev->dev_addr);
-
err = register_netdev(dev);
if (!err)
return 0;
+ netdev_info(dev, "MAC %pM\n", dev->dev_addr);
+
free_netdev(dev);
return err;
}
@@ -589,7 +588,6 @@ static irqreturn_t mace_interrupt(int irq, void *dev_id)
else if (fs & (UFLO|LCOL|RTRY)) {
++dev->stats.tx_aborted_errors;
if (mb->xmtfs & UFLO) {
- printk(KERN_ERR "%s: DMA underrun.\n", dev->name);
dev->stats.tx_fifo_errors++;
mace_txdma_reset(dev);
}
@@ -644,10 +642,8 @@ static void mace_dma_rx_frame(struct net_device *dev, struct mace_frame *mf)
if (frame_status & (RS_OFLO | RS_CLSN | RS_FRAMERR | RS_FCSERR)) {
dev->stats.rx_errors++;
- if (frame_status & RS_OFLO) {
- printk(KERN_DEBUG "%s: fifo overflow.\n", dev->name);
+ if (frame_status & RS_OFLO)
dev->stats.rx_fifo_errors++;
- }
if (frame_status & RS_CLSN)
dev->stats.collisions++;
if (frame_status & RS_FRAMERR)
--
2.13.5
^ permalink raw reply related
* [PATCH RESEND net 9/9] net/mac8390: Fix log messages
From: Finn Thain @ 2017-10-03 1:07 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, linux-kernel
In-Reply-To: <cover.1506992619.git.fthain@telegraphics.com.au>
Before expansion, dev->name is "eth%d", so log the slot number instead.
Log the MAC address after the interface gets a meaningful name.
Disambiguate the two identical "Card type %s is unsupported" messages.
Fix the duplicated driver name in the pr_warn() message.
Fixes: 3a3a7f3b7fbd ("net: mac8390: Allow modular build")
Signed-off-by: Finn Thain <fthain@telegraphics.com.au>
---
drivers/net/ethernet/8390/mac8390.c | 35 ++++++++++++++++++-----------------
1 file changed, 18 insertions(+), 17 deletions(-)
diff --git a/drivers/net/ethernet/8390/mac8390.c b/drivers/net/ethernet/8390/mac8390.c
index 1bfc66f37971..45d724d8333a 100644
--- a/drivers/net/ethernet/8390/mac8390.c
+++ b/drivers/net/ethernet/8390/mac8390.c
@@ -307,14 +307,15 @@ static bool __init mac8390_init(struct net_device *dev, struct nubus_dev *ndev,
*/
if (nubus_get_func_dir(ndev, &dir) == -1) {
- pr_err("%s: Unable to get Nubus functional directory for slot %X!\n",
- dev->name, ndev->board->slot);
+ pr_err("Unable to get Nubus functional directory for slot %X!\n",
+ ndev->board->slot);
return false;
}
/* Get the MAC address */
if (nubus_find_rsrc(&dir, NUBUS_RESID_MAC_ADDRESS, &ent) == -1) {
- pr_info("%s: Couldn't get MAC address!\n", dev->name);
+ pr_info("MAC address resource for slot %X not found!\n",
+ ndev->board->slot);
return false;
}
@@ -324,8 +325,8 @@ static bool __init mac8390_init(struct net_device *dev, struct nubus_dev *ndev,
nubus_rewinddir(&dir);
if (nubus_find_rsrc(&dir, NUBUS_RESID_MINOR_BASEOS,
&ent) == -1) {
- pr_err("%s: Memory offset resource for slot %X not found!\n",
- dev->name, ndev->board->slot);
+ pr_err("Memory offset resource for slot %X not found!\n",
+ ndev->board->slot);
return false;
}
nubus_get_rsrc_mem(&offset, &ent, 4);
@@ -335,8 +336,8 @@ static bool __init mac8390_init(struct net_device *dev, struct nubus_dev *ndev,
nubus_rewinddir(&dir);
if (nubus_find_rsrc(&dir, NUBUS_RESID_MINOR_LENGTH,
&ent) == -1) {
- pr_info("%s: Memory length resource for slot %X not found, probing\n",
- dev->name, ndev->board->slot);
+ pr_info("Memory length resource for slot %X not found, probing\n",
+ ndev->board->slot);
offset = mac8390_memsize(dev->mem_start);
} else {
nubus_get_rsrc_mem(&offset, &ent, 4);
@@ -379,8 +380,8 @@ static bool __init mac8390_init(struct net_device *dev, struct nubus_dev *ndev,
break;
default:
- pr_err("Card type %s is unsupported, sorry\n",
- ndev->board->name);
+ pr_err("No known base address for %s card in slot %X\n",
+ ndev->board->name, ndev->board->slot);
return false;
}
}
@@ -435,6 +436,9 @@ struct net_device * __init mac8390_probe(int unit)
err = register_netdev(dev);
if (err)
goto out;
+
+ netdev_info(dev, "MAC %pM, IRQ %d\n", dev->dev_addr, dev->irq);
+
return dev;
out:
@@ -453,7 +457,7 @@ int __init init_module(void)
{
dev_mac8390 = mac8390_probe(-1);
if (IS_ERR(dev_mac8390)) {
- pr_warn("mac8390: No card found\n");
+ pr_warn("No card found\n");
return PTR_ERR(dev_mac8390);
}
return 0;
@@ -600,19 +604,16 @@ static int __init mac8390_initdev(struct net_device *dev,
break;
default:
- pr_err("Card type %s is unsupported, sorry\n",
- ndev->board->name);
+ pr_err("%s card in slot %X is unsupported, sorry\n",
+ ndev->board->name, ndev->board->slot);
return -ENODEV;
}
__NS8390_init(dev, 0);
/* Good, done, now spit out some messages */
- pr_info("%s: %s in slot %X (type %s)\n",
- dev->name, ndev->board->name, ndev->board->slot,
- cardname[type]);
- pr_info("MAC %pM IRQ %d, %d KB shared memory at %#lx, %d-bit access.\n",
- dev->dev_addr, dev->irq,
+ pr_info("%s in slot %X has type %s, %d KB shared memory at %#lx, %d-bit access\n",
+ ndev->board->name, ndev->board->slot, cardname[type],
(unsigned int)(dev->mem_end - dev->mem_start) >> 10,
dev->mem_start, access_bitmode ? 32 : 16);
return 0;
--
2.13.5
^ permalink raw reply related
* [PATCH RESEND net 7/9] net/sonic: Replace custom debug logging with netif_* calls
From: Finn Thain @ 2017-10-03 1:07 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, linux-kernel, Thomas Bogendoerfer, Chris Zankel
In-Reply-To: <cover.1506992619.git.fthain@telegraphics.com.au>
Also eliminate duplicated debug code by moving it into the core driver.
Cc: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
Cc: Chris Zankel <chris@zankel.net>
Signed-off-by: Finn Thain <fthain@telegraphics.com.au>
---
drivers/net/ethernet/natsemi/jazzsonic.c | 17 ++----
drivers/net/ethernet/natsemi/macsonic.c | 21 +-------
drivers/net/ethernet/natsemi/sonic.c | 92 +++++++++++++++-----------------
drivers/net/ethernet/natsemi/sonic.h | 2 +
drivers/net/ethernet/natsemi/xtsonic.c | 17 ++----
5 files changed, 54 insertions(+), 95 deletions(-)
diff --git a/drivers/net/ethernet/natsemi/jazzsonic.c b/drivers/net/ethernet/natsemi/jazzsonic.c
index 6ec25c27cb73..0fc303901f27 100644
--- a/drivers/net/ethernet/natsemi/jazzsonic.c
+++ b/drivers/net/ethernet/natsemi/jazzsonic.c
@@ -59,14 +59,6 @@ do { \
*((volatile unsigned int *)dev->base_addr+(reg)) = (val); \
} while (0)
-
-/* use 0 for production, 1 for verification, >1 for debug */
-#ifdef SONIC_DEBUG
-static unsigned int sonic_debug = SONIC_DEBUG;
-#else
-static unsigned int sonic_debug = 1;
-#endif
-
/*
* We cannot use station (ethernet) address prefixes to detect the
* sonic controller since these are board manufacturer depended.
@@ -116,7 +108,6 @@ static const struct net_device_ops sonic_netdev_ops = {
static int sonic_probe1(struct net_device *dev)
{
- static unsigned version_printed;
unsigned int silicon_revision;
unsigned int val;
struct sonic_local *lp = netdev_priv(dev);
@@ -143,9 +134,6 @@ static int sonic_probe1(struct net_device *dev)
goto out;
}
- if (sonic_debug && version_printed++ == 0)
- printk(version);
-
printk(KERN_INFO "%s: Sonic ethernet found at 0x%08lx, ",
dev_name(lp->device), dev->base_addr);
@@ -241,6 +229,9 @@ static int jazz_sonic_probe(struct platform_device *pdev)
err = sonic_probe1(dev);
if (err)
goto out;
+
+ sonic_msg_init(dev);
+
err = register_netdev(dev);
if (err)
goto out1;
@@ -258,8 +249,6 @@ static int jazz_sonic_probe(struct platform_device *pdev)
}
MODULE_DESCRIPTION("Jazz SONIC ethernet driver");
-module_param(sonic_debug, int, 0);
-MODULE_PARM_DESC(sonic_debug, "jazzsonic debug level (1-4)");
MODULE_ALIAS("platform:jazzsonic");
#include "sonic.c"
diff --git a/drivers/net/ethernet/natsemi/macsonic.c b/drivers/net/ethernet/natsemi/macsonic.c
index 795d71522617..6e5f87e955da 100644
--- a/drivers/net/ethernet/natsemi/macsonic.c
+++ b/drivers/net/ethernet/natsemi/macsonic.c
@@ -71,15 +71,6 @@ static char mac_sonic_string[] = "macsonic";
#define SONIC_WRITE(reg,val) (nubus_writew(val, dev->base_addr + (reg * 4) \
+ lp->reg_offset))
-/* use 0 for production, 1 for verification, >1 for debug */
-#ifdef SONIC_DEBUG
-static unsigned int sonic_debug = SONIC_DEBUG;
-#else
-static unsigned int sonic_debug = 1;
-#endif
-
-static int sonic_version_printed;
-
/* For onboard SONIC */
#define ONBOARD_SONIC_REGISTERS 0x50F0A000
#define ONBOARD_SONIC_PROM_BASE 0x50f08000
@@ -341,10 +332,6 @@ static int mac_onboard_sonic_probe(struct net_device *dev)
else
dev->irq = IRQ_NUBUS_9;
- if (!sonic_version_printed) {
- printk(KERN_INFO "%s", version);
- sonic_version_printed = 1;
- }
printk(KERN_INFO "%s: onboard / comm-slot SONIC at 0x%08lx\n",
dev_name(lp->device), dev->base_addr);
@@ -527,10 +514,6 @@ static int mac_nubus_sonic_probe(struct net_device *dev)
lp->dma_bitmode = dma_bitmode;
dev->irq = SLOT2IRQ(ndev->board->slot);
- if (!sonic_version_printed) {
- printk(KERN_INFO "%s", version);
- sonic_version_printed = 1;
- }
printk(KERN_INFO "%s: %s in slot %X\n",
dev_name(lp->device), ndev->board->name, ndev->board->slot);
printk(KERN_INFO "%s: revision 0x%04x, using %d bit DMA and register offset %d\n",
@@ -586,6 +569,8 @@ static int mac_sonic_probe(struct platform_device *pdev)
if (err)
goto out;
found:
+ sonic_msg_init(dev);
+
err = register_netdev(dev);
if (err)
goto out;
@@ -601,8 +586,6 @@ static int mac_sonic_probe(struct platform_device *pdev)
}
MODULE_DESCRIPTION("Macintosh SONIC ethernet driver");
-module_param(sonic_debug, int, 0);
-MODULE_PARM_DESC(sonic_debug, "macsonic debug level (1-4)");
MODULE_ALIAS("platform:macsonic");
#include "sonic.c"
diff --git a/drivers/net/ethernet/natsemi/sonic.c b/drivers/net/ethernet/natsemi/sonic.c
index 612c7a44b26c..c0bae615238f 100644
--- a/drivers/net/ethernet/natsemi/sonic.c
+++ b/drivers/net/ethernet/natsemi/sonic.c
@@ -33,7 +33,18 @@
* the NetBSD file "sys/arch/mac68k/dev/if_sn.c".
*/
+static int sonic_debug = -1;
+module_param(sonic_debug, int, 0);
+MODULE_PARM_DESC(sonic_debug, "debug message level");
+static void sonic_msg_init(struct net_device *dev)
+{
+ struct sonic_local *lp = netdev_priv(dev);
+
+ lp->msg_enable = netif_msg_init(sonic_debug, 0);
+
+ netif_dbg(lp, drv, dev, "%s", version);
+}
/*
* Open/initialize the SONIC controller.
@@ -47,8 +58,7 @@ static int sonic_open(struct net_device *dev)
struct sonic_local *lp = netdev_priv(dev);
int i;
- if (sonic_debug > 2)
- printk("sonic_open: initializing sonic driver.\n");
+ netif_dbg(lp, ifup, dev, "%s\n", __func__);
for (i = 0; i < SONIC_NUM_RRS; i++) {
struct sk_buff *skb = netdev_alloc_skb(dev, SONIC_RBSIZE + 2);
@@ -95,8 +105,7 @@ static int sonic_open(struct net_device *dev)
netif_start_queue(dev);
- if (sonic_debug > 2)
- printk("sonic_open: Initialization done.\n");
+ netif_dbg(lp, ifup, dev, "%s: done\n", __func__);
return 0;
}
@@ -110,8 +119,7 @@ static int sonic_close(struct net_device *dev)
struct sonic_local *lp = netdev_priv(dev);
int i;
- if (sonic_debug > 2)
- printk("sonic_close\n");
+ netif_dbg(lp, ifdown, dev, "%s\n", __func__);
netif_stop_queue(dev);
@@ -205,8 +213,7 @@ static int sonic_send_packet(struct sk_buff *skb, struct net_device *dev)
int length;
int entry = lp->next_tx;
- if (sonic_debug > 2)
- printk("sonic_send_packet: skb=%p, dev=%p\n", skb, dev);
+ netif_dbg(lp, tx_queued, dev, "%s: skb=%p\n", __func__, skb);
length = skb->len;
if (length < ETH_ZLEN) {
@@ -252,14 +259,12 @@ static int sonic_send_packet(struct sk_buff *skb, struct net_device *dev)
lp->next_tx = (entry + 1) & SONIC_TDS_MASK;
if (lp->tx_skb[lp->next_tx] != NULL) {
/* The ring is full, the ISR has yet to process the next TD. */
- if (sonic_debug > 3)
- printk("%s: stopping queue\n", dev->name);
+ netif_dbg(lp, tx_queued, dev, "%s: stop queue\n", __func__);
netif_stop_queue(dev);
/* after this packet, wait for ISR to free up some TDAs */
} else netif_start_queue(dev);
- if (sonic_debug > 2)
- printk("sonic_send_packet: issuing Tx command\n");
+ netif_dbg(lp, tx_queued, dev, "%s: issuing tx command\n", __func__);
SONIC_WRITE(SONIC_CMD, SONIC_CR_TXP);
@@ -281,8 +286,7 @@ static irqreturn_t sonic_interrupt(int irq, void *dev_id)
do {
if (status & SONIC_INT_PKTRX) {
- if (sonic_debug > 2)
- printk("%s: packet rx\n", dev->name);
+ netif_dbg(lp, intr, dev, "%s: packet rx\n", __func__);
sonic_rx(dev); /* got packet(s) */
SONIC_WRITE(SONIC_ISR, SONIC_INT_PKTRX); /* clear the interrupt */
}
@@ -299,8 +303,7 @@ static irqreturn_t sonic_interrupt(int irq, void *dev_id)
* still being allocated by sonic_send_packet (status clear & tx_skb[entry] clear)
*/
- if (sonic_debug > 2)
- printk("%s: tx done\n", dev->name);
+ netif_dbg(lp, intr, dev, "%s: tx done\n", __func__);
while (lp->tx_skb[entry] != NULL) {
if ((td_status = sonic_tda_get(dev, entry, SONIC_TD_STATUS)) == 0)
@@ -346,20 +349,20 @@ static irqreturn_t sonic_interrupt(int irq, void *dev_id)
* check error conditions
*/
if (status & SONIC_INT_RFO) {
- if (sonic_debug > 1)
- printk("%s: rx fifo overrun\n", dev->name);
+ netif_dbg(lp, rx_err, dev, "%s: rx fifo overrun\n",
+ __func__);
lp->stats.rx_fifo_errors++;
SONIC_WRITE(SONIC_ISR, SONIC_INT_RFO); /* clear the interrupt */
}
if (status & SONIC_INT_RDE) {
- if (sonic_debug > 1)
- printk("%s: rx descriptors exhausted\n", dev->name);
+ netif_dbg(lp, rx_err, dev, "%s: rx descriptors exhausted\n",
+ __func__);
lp->stats.rx_dropped++;
SONIC_WRITE(SONIC_ISR, SONIC_INT_RDE); /* clear the interrupt */
}
if (status & SONIC_INT_RBAE) {
- if (sonic_debug > 1)
- printk("%s: rx buffer area exceeded\n", dev->name);
+ netif_dbg(lp, rx_err, dev, "%s: rx buffer area exceeded\n",
+ __func__);
lp->stats.rx_dropped++;
SONIC_WRITE(SONIC_ISR, SONIC_INT_RBAE); /* clear the interrupt */
}
@@ -380,8 +383,9 @@ static irqreturn_t sonic_interrupt(int irq, void *dev_id)
/* transmit error */
if (status & SONIC_INT_TXER) {
- if ((SONIC_READ(SONIC_TCR) & SONIC_TCR_FU) && (sonic_debug > 2))
- printk(KERN_ERR "%s: tx fifo underrun\n", dev->name);
+ if (SONIC_READ(SONIC_TCR) & SONIC_TCR_FU)
+ netif_dbg(lp, tx_err, dev, "%s: tx fifo underrun\n",
+ __func__);
SONIC_WRITE(SONIC_ISR, SONIC_INT_TXER); /* clear the interrupt */
}
@@ -475,8 +479,8 @@ static void sonic_rx(struct net_device *dev)
if (lp->cur_rwp >= lp->rra_end) lp->cur_rwp = lp->rra_laddr & 0xffff;
SONIC_WRITE(SONIC_RWP, lp->cur_rwp);
if (SONIC_READ(SONIC_ISR) & SONIC_INT_RBE) {
- if (sonic_debug > 2)
- printk("%s: rx buffer exhausted\n", dev->name);
+ netif_dbg(lp, rx_err, dev, "%s: rx buffer exhausted\n",
+ __func__);
SONIC_WRITE(SONIC_ISR, SONIC_INT_RBE); /* clear the flag */
}
} else
@@ -542,9 +546,8 @@ static void sonic_multicast_list(struct net_device *dev)
(netdev_mc_count(dev) > 15)) {
rcr |= SONIC_RCR_AMC;
} else {
- if (sonic_debug > 2)
- printk("sonic_multicast_list: mc_count %d\n",
- netdev_mc_count(dev));
+ netif_dbg(lp, ifup, dev, "%s: mc_count %d\n", __func__,
+ netdev_mc_count(dev));
sonic_set_cam_enable(dev, 1); /* always enable our own address */
i = 1;
netdev_for_each_mc_addr(ha, dev) {
@@ -562,8 +565,7 @@ static void sonic_multicast_list(struct net_device *dev)
}
}
- if (sonic_debug > 2)
- printk("sonic_multicast_list: setting RCR=%x\n", rcr);
+ netif_dbg(lp, ifup, dev, "%s: rcr=%x\n", __func__, rcr);
SONIC_WRITE(SONIC_RCR, rcr);
}
@@ -596,8 +598,7 @@ static int sonic_init(struct net_device *dev)
/*
* initialize the receive resource area
*/
- if (sonic_debug > 2)
- printk("sonic_init: initialize receive resource area\n");
+ netif_dbg(lp, ifup, dev, "%s: receive resource area\n", __func__);
for (i = 0; i < SONIC_NUM_RRS; i++) {
u16 bufadr_l = (unsigned long)lp->rx_laddr[i] & 0xffff;
@@ -622,8 +623,7 @@ static int sonic_init(struct net_device *dev)
SONIC_WRITE(SONIC_EOBC, (SONIC_RBSIZE >> 1) - (lp->dma_bitmode ? 2 : 1));
/* load the resource pointers */
- if (sonic_debug > 3)
- printk("sonic_init: issuing RRRA command\n");
+ netif_dbg(lp, ifup, dev, "%s: issuing RRRA command\n", __func__);
SONIC_WRITE(SONIC_CMD, SONIC_CR_RRRA);
i = 0;
@@ -632,16 +632,16 @@ static int sonic_init(struct net_device *dev)
break;
}
- if (sonic_debug > 2)
- printk("sonic_init: status=%x i=%d\n", SONIC_READ(SONIC_CMD), i);
+ netif_dbg(lp, ifup, dev, "%s: status=%x, i=%d\n", __func__,
+ SONIC_READ(SONIC_CMD), i);
/*
* Initialize the receive descriptors so that they
* become a circular linked list, ie. let the last
* descriptor point to the first again.
*/
- if (sonic_debug > 2)
- printk("sonic_init: initialize receive descriptors\n");
+ netif_dbg(lp, ifup, dev, "%s: receive descriptors\n", __func__);
+
for (i=0; i<SONIC_NUM_RDS; i++) {
sonic_rda_put(dev, i, SONIC_RD_STATUS, 0);
sonic_rda_put(dev, i, SONIC_RD_PKTLEN, 0);
@@ -664,8 +664,8 @@ static int sonic_init(struct net_device *dev)
/*
* initialize transmit descriptors
*/
- if (sonic_debug > 2)
- printk("sonic_init: initialize transmit descriptors\n");
+ netif_dbg(lp, ifup, dev, "%s: transmit descriptors\n", __func__);
+
for (i = 0; i < SONIC_NUM_TDS; i++) {
sonic_tda_put(dev, i, SONIC_TD_STATUS, 0);
sonic_tda_put(dev, i, SONIC_TD_CONFIG, 0);
@@ -712,10 +712,8 @@ static int sonic_init(struct net_device *dev)
if (SONIC_READ(SONIC_ISR) & SONIC_INT_LCD)
break;
}
- if (sonic_debug > 2) {
- printk("sonic_init: CMD=%x, ISR=%x\n, i=%d",
- SONIC_READ(SONIC_CMD), SONIC_READ(SONIC_ISR), i);
- }
+ netif_dbg(lp, ifup, dev, "%s: CMD=%x, ISR=%x, i=%d\n", __func__,
+ SONIC_READ(SONIC_CMD), SONIC_READ(SONIC_ISR), i);
/*
* enable receiver, disable loopback
@@ -731,9 +729,7 @@ static int sonic_init(struct net_device *dev)
if ((cmd & SONIC_CR_RXEN) == 0 || (cmd & SONIC_CR_STP) == 0)
printk(KERN_ERR "sonic_init: failed, status=%x\n", cmd);
- if (sonic_debug > 2)
- printk("sonic_init: new status=%x\n",
- SONIC_READ(SONIC_CMD));
+ netif_dbg(lp, ifup, dev, "%s: cmd=%x\n", __func__, cmd);
return 0;
}
diff --git a/drivers/net/ethernet/natsemi/sonic.h b/drivers/net/ethernet/natsemi/sonic.h
index 7b0a8db57af9..2f18229b17cf 100644
--- a/drivers/net/ethernet/natsemi/sonic.h
+++ b/drivers/net/ethernet/natsemi/sonic.h
@@ -318,6 +318,7 @@ struct sonic_local {
unsigned int eol_rx;
unsigned int eol_tx; /* last unacked transmit packet */
unsigned int next_tx; /* next free TD */
+ int msg_enable;
struct device *device; /* generic device */
struct net_device_stats stats;
};
@@ -335,6 +336,7 @@ static struct net_device_stats *sonic_get_stats(struct net_device *dev);
static void sonic_multicast_list(struct net_device *dev);
static int sonic_init(struct net_device *dev);
static void sonic_tx_timeout(struct net_device *dev);
+static void sonic_msg_init(struct net_device *dev);
/* Internal inlines for reading/writing DMA buffers. Note that bus
size and endianness matter here, whereas they don't for registers,
diff --git a/drivers/net/ethernet/natsemi/xtsonic.c b/drivers/net/ethernet/natsemi/xtsonic.c
index bed89e8105e9..1f601eb8a828 100644
--- a/drivers/net/ethernet/natsemi/xtsonic.c
+++ b/drivers/net/ethernet/natsemi/xtsonic.c
@@ -72,14 +72,6 @@ extern void xtboard_get_ether_addr(unsigned char *buf);
#define SONIC_WRITE(reg,val) \
*((volatile unsigned int *)dev->base_addr+reg) = val
-
-/* Use 0 for production, 1 for verification, and >2 for debug */
-#ifdef SONIC_DEBUG
-static unsigned int sonic_debug = SONIC_DEBUG;
-#else
-static unsigned int sonic_debug = 1;
-#endif
-
/*
* We cannot use station (ethernet) address prefixes to detect the
* sonic controller since these are board manufacturer depended.
@@ -129,7 +121,6 @@ static const struct net_device_ops xtsonic_netdev_ops = {
static int __init sonic_probe1(struct net_device *dev)
{
- static unsigned version_printed = 0;
unsigned int silicon_revision;
struct sonic_local *lp = netdev_priv(dev);
unsigned int base_addr = dev->base_addr;
@@ -156,9 +147,6 @@ static int __init sonic_probe1(struct net_device *dev)
return -ENODEV;
}
- if (sonic_debug && version_printed++ == 0)
- printk(version);
-
/*
* Put the sonic into software reset, then retrieve ethernet address.
* Note: we are assuming that the boot-loader has initialized the cam.
@@ -269,6 +257,9 @@ int xtsonic_probe(struct platform_device *pdev)
if ((err = sonic_probe1(dev)))
goto out;
+
+ sonic_msg_init(dev);
+
if ((err = register_netdev(dev)))
goto out1;
@@ -286,8 +277,6 @@ int xtsonic_probe(struct platform_device *pdev)
}
MODULE_DESCRIPTION("Xtensa XT2000 SONIC ethernet driver");
-module_param(sonic_debug, int, 0);
-MODULE_PARM_DESC(sonic_debug, "xtsonic debug level (1-4)");
#include "sonic.c"
--
2.13.5
^ permalink raw reply related
* [PATCH RESEND net 8/9] net/8390: Fix redundant code
From: Finn Thain @ 2017-10-03 1:07 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, linux-kernel, Russell King, linux-arm-kernel
In-Reply-To: <cover.1506992619.git.fthain@telegraphics.com.au>
The patch which introduced the 8390 core module parameter 'msg_enable'
failed to do anything useful with it: it merely causes an ancient
version string to be logged.
Remove the other code that logs the same string. Use the msg_enable
module parameter as the default value for ei_local->msg_enable.
Otherwise, some 8390 modules have no way to set ei_local->msg_enable.
Also fix two more issues arising from the same patch: indentation
mistakes and pointless static variables.
Fixes: c45f812f0280 ("8390 : Replace ei_debug with msg_enable/NETIF_MSG_* feature")
Cc: Russell King <linux@armlinux.org.uk>
Cc: linux-arm-kernel@lists.infradead.org
Signed-off-by: Finn Thain <fthain@telegraphics.com.au>
---
drivers/net/ethernet/8390/ax88796.c | 3 ---
drivers/net/ethernet/8390/axnet_cs.c | 2 --
drivers/net/ethernet/8390/etherh.c | 17 -----------------
drivers/net/ethernet/8390/hydra.c | 4 ----
drivers/net/ethernet/8390/lib8390.c | 2 ++
drivers/net/ethernet/8390/mac8390.c | 7 -------
drivers/net/ethernet/8390/mcf8390.c | 4 ----
drivers/net/ethernet/8390/pcnet_cs.c | 4 ----
drivers/net/ethernet/8390/zorro8390.c | 5 -----
9 files changed, 2 insertions(+), 46 deletions(-)
diff --git a/drivers/net/ethernet/8390/ax88796.c b/drivers/net/ethernet/8390/ax88796.c
index 05d9d3e2e92e..28aa79d2f16c 100644
--- a/drivers/net/ethernet/8390/ax88796.c
+++ b/drivers/net/ethernet/8390/ax88796.c
@@ -77,8 +77,6 @@ static unsigned char version[] = "ax88796.c: Copyright 2005,2007 Simtec Electron
#define AX_GPOC_PPDSET BIT(6)
-static u32 ax_msg_enable;
-
/* device private data */
struct ax_device {
@@ -747,7 +745,6 @@ static int ax_init_dev(struct net_device *dev)
ei_local->block_output = &ax_block_output;
ei_local->get_8390_hdr = &ax_get_8390_hdr;
ei_local->priv = 0;
- ei_local->msg_enable = ax_msg_enable;
dev->netdev_ops = &ax_netdev_ops;
dev->ethtool_ops = &ax_ethtool_ops;
diff --git a/drivers/net/ethernet/8390/axnet_cs.c b/drivers/net/ethernet/8390/axnet_cs.c
index 3da1fc539ef9..91e76dc1e6e1 100644
--- a/drivers/net/ethernet/8390/axnet_cs.c
+++ b/drivers/net/ethernet/8390/axnet_cs.c
@@ -104,7 +104,6 @@ static void AX88190_init(struct net_device *dev, int startp);
static int ax_open(struct net_device *dev);
static int ax_close(struct net_device *dev);
static irqreturn_t ax_interrupt(int irq, void *dev_id);
-static u32 axnet_msg_enable;
/*====================================================================*/
@@ -151,7 +150,6 @@ static int axnet_probe(struct pcmcia_device *link)
return -ENOMEM;
ei_local = netdev_priv(dev);
- ei_local->msg_enable = axnet_msg_enable;
spin_lock_init(&ei_local->page_lock);
info = PRIV(dev);
diff --git a/drivers/net/ethernet/8390/etherh.c b/drivers/net/ethernet/8390/etherh.c
index 11cbf22ad201..32e9627e3880 100644
--- a/drivers/net/ethernet/8390/etherh.c
+++ b/drivers/net/ethernet/8390/etherh.c
@@ -64,8 +64,6 @@ static char version[] =
#include "lib8390.c"
-static u32 etherh_msg_enable;
-
struct etherh_priv {
void __iomem *ioc_fast;
void __iomem *memc;
@@ -502,18 +500,6 @@ etherh_close(struct net_device *dev)
}
/*
- * Initialisation
- */
-
-static void __init etherh_banner(void)
-{
- static int version_printed;
-
- if ((etherh_msg_enable & NETIF_MSG_DRV) && (version_printed++ == 0))
- pr_info("%s", version);
-}
-
-/*
* Read the ethernet address string from the on board rom.
* This is an ascii string...
*/
@@ -671,8 +657,6 @@ etherh_probe(struct expansion_card *ec, const struct ecard_id *id)
struct etherh_priv *eh;
int ret;
- etherh_banner();
-
ret = ecard_request_resources(ec);
if (ret)
goto out;
@@ -757,7 +741,6 @@ etherh_probe(struct expansion_card *ec, const struct ecard_id *id)
ei_local->block_output = etherh_block_output;
ei_local->get_8390_hdr = etherh_get_header;
ei_local->interface_num = 0;
- ei_local->msg_enable = etherh_msg_enable;
etherh_reset(dev);
__NS8390_init(dev, 0);
diff --git a/drivers/net/ethernet/8390/hydra.c b/drivers/net/ethernet/8390/hydra.c
index 8ae249195301..941754ea78ec 100644
--- a/drivers/net/ethernet/8390/hydra.c
+++ b/drivers/net/ethernet/8390/hydra.c
@@ -66,7 +66,6 @@ static void hydra_block_input(struct net_device *dev, int count,
static void hydra_block_output(struct net_device *dev, int count,
const unsigned char *buf, int start_page);
static void hydra_remove_one(struct zorro_dev *z);
-static u32 hydra_msg_enable;
static struct zorro_device_id hydra_zorro_tbl[] = {
{ ZORRO_PROD_HYDRA_SYSTEMS_AMIGANET },
@@ -119,7 +118,6 @@ static int hydra_init(struct zorro_dev *z)
int start_page, stop_page;
int j;
int err;
- struct ei_device *ei_local;
static u32 hydra_offsets[16] = {
0x00, 0x02, 0x04, 0x06, 0x08, 0x0a, 0x0c, 0x0e,
@@ -138,8 +136,6 @@ static int hydra_init(struct zorro_dev *z)
start_page = NESM_START_PG;
stop_page = NESM_STOP_PG;
- ei_local = netdev_priv(dev);
- ei_local->msg_enable = hydra_msg_enable;
dev->base_addr = ioaddr;
dev->irq = IRQ_AMIGA_PORTS;
diff --git a/drivers/net/ethernet/8390/lib8390.c b/drivers/net/ethernet/8390/lib8390.c
index 60f8e2c8e726..5d9bbde9fe68 100644
--- a/drivers/net/ethernet/8390/lib8390.c
+++ b/drivers/net/ethernet/8390/lib8390.c
@@ -975,6 +975,8 @@ static void ethdev_setup(struct net_device *dev)
ether_setup(dev);
spin_lock_init(&ei_local->page_lock);
+
+ ei_local->msg_enable = msg_enable;
}
/**
diff --git a/drivers/net/ethernet/8390/mac8390.c b/drivers/net/ethernet/8390/mac8390.c
index 9497f18eaba0..1bfc66f37971 100644
--- a/drivers/net/ethernet/8390/mac8390.c
+++ b/drivers/net/ethernet/8390/mac8390.c
@@ -167,7 +167,6 @@ static void slow_sane_block_output(struct net_device *dev, int count,
const unsigned char *buf, int start_page);
static void word_memcpy_tocard(unsigned long tp, const void *fp, int count);
static void word_memcpy_fromcard(void *tp, unsigned long fp, int count);
-static u32 mac8390_msg_enable;
static enum mac8390_type __init mac8390_ident(struct nubus_dev *dev)
{
@@ -297,8 +296,6 @@ static bool __init mac8390_init(struct net_device *dev, struct nubus_dev *ndev,
int offset;
volatile unsigned short *i;
- printk_once(KERN_INFO pr_fmt("%s"), version);
-
dev->irq = SLOT2IRQ(ndev->board->slot);
/* This is getting to be a habit */
dev->base_addr = (ndev->board->slot_addr |
@@ -396,7 +393,6 @@ struct net_device * __init mac8390_probe(int unit)
struct net_device *dev;
struct nubus_dev *ndev = NULL;
int err = -ENODEV;
- struct ei_device *ei_local;
static unsigned int slots;
@@ -436,9 +432,6 @@ struct net_device * __init mac8390_probe(int unit)
if (!ndev)
goto out;
- ei_local = netdev_priv(dev);
- ei_local->msg_enable = mac8390_msg_enable;
-
err = register_netdev(dev);
if (err)
goto out;
diff --git a/drivers/net/ethernet/8390/mcf8390.c b/drivers/net/ethernet/8390/mcf8390.c
index 4bb967bc879e..4ad8031ab669 100644
--- a/drivers/net/ethernet/8390/mcf8390.c
+++ b/drivers/net/ethernet/8390/mcf8390.c
@@ -38,7 +38,6 @@ static const char version[] =
#define NESM_START_PG 0x40 /* First page of TX buffer */
#define NESM_STOP_PG 0x80 /* Last page +1 of RX ring */
-static u32 mcf8390_msg_enable;
#ifdef NE2000_ODDOFFSET
/*
@@ -407,7 +406,6 @@ static int mcf8390_init(struct net_device *dev)
static int mcf8390_probe(struct platform_device *pdev)
{
struct net_device *dev;
- struct ei_device *ei_local;
struct resource *mem, *irq;
resource_size_t msize;
int ret;
@@ -435,8 +433,6 @@ static int mcf8390_probe(struct platform_device *pdev)
SET_NETDEV_DEV(dev, &pdev->dev);
platform_set_drvdata(pdev, dev);
- ei_local = netdev_priv(dev);
- ei_local->msg_enable = mcf8390_msg_enable;
dev->irq = irq->start;
dev->base_addr = mem->start;
diff --git a/drivers/net/ethernet/8390/pcnet_cs.c b/drivers/net/ethernet/8390/pcnet_cs.c
index bd0a2a14b649..a81ffe4874e1 100644
--- a/drivers/net/ethernet/8390/pcnet_cs.c
+++ b/drivers/net/ethernet/8390/pcnet_cs.c
@@ -66,7 +66,6 @@
#define PCNET_RDC_TIMEOUT (2*HZ/100) /* Max wait in jiffies for Tx RDC */
static const char *if_names[] = { "auto", "10baseT", "10base2"};
-static u32 pcnet_msg_enable;
/*====================================================================*/
@@ -556,7 +555,6 @@ static int pcnet_config(struct pcmcia_device *link)
int start_pg, stop_pg, cm_offset;
int has_shmem = 0;
struct hw_info *local_hw_info;
- struct ei_device *ei_local;
dev_dbg(&link->dev, "pcnet_config\n");
@@ -606,8 +604,6 @@ static int pcnet_config(struct pcmcia_device *link)
mii_phy_probe(dev);
SET_NETDEV_DEV(dev, &link->dev);
- ei_local = netdev_priv(dev);
- ei_local->msg_enable = pcnet_msg_enable;
if (register_netdev(dev) != 0) {
pr_notice("register_netdev() failed\n");
diff --git a/drivers/net/ethernet/8390/zorro8390.c b/drivers/net/ethernet/8390/zorro8390.c
index 6d93956b293b..35a500a21521 100644
--- a/drivers/net/ethernet/8390/zorro8390.c
+++ b/drivers/net/ethernet/8390/zorro8390.c
@@ -44,8 +44,6 @@
static const char version[] =
"8390.c:v1.10cvs 9/23/94 Donald Becker (becker@cesdis.gsfc.nasa.gov)\n";
-static u32 zorro8390_msg_enable;
-
#include "lib8390.c"
#define DRV_NAME "zorro8390"
@@ -296,7 +294,6 @@ static int zorro8390_init(struct net_device *dev, unsigned long board,
int err;
unsigned char SA_prom[32];
int start_page, stop_page;
- struct ei_device *ei_local = netdev_priv(dev);
static u32 zorro8390_offsets[16] = {
0x00, 0x02, 0x04, 0x06, 0x08, 0x0a, 0x0c, 0x0e,
0x10, 0x12, 0x14, 0x16, 0x18, 0x1a, 0x1c, 0x1e,
@@ -388,8 +385,6 @@ static int zorro8390_init(struct net_device *dev, unsigned long board,
dev->netdev_ops = &zorro8390_netdev_ops;
__NS8390_init(dev, 0);
- ei_local->msg_enable = zorro8390_msg_enable;
-
err = register_netdev(dev);
if (err) {
free_irq(IRQ_AMIGA_PORTS, dev);
--
2.13.5
^ permalink raw reply related
* [PATCH RESEND net 1/9] net/smc9194: Remove bogus CONFIG_MAC reference
From: Finn Thain @ 2017-10-03 1:07 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, linux-kernel
In-Reply-To: <cover.1506992619.git.fthain@telegraphics.com.au>
The only version of smc9194.c with Mac support is the one in the
linux-mac68k CVS repo. AFAIK that driver never made it to the mainline.
Despite that, as of v2.3.45, arch/m68k/config.in listed CONFIG_SMC9194
under CONFIG_MAC. This mistake got carried over into Kconfig in v2.5.55.
(See pre-git era "[PATCH] add m68k dependencies to net driver config".)
Signed-off-by: Finn Thain <fthain@telegraphics.com.au>
---
drivers/net/ethernet/smsc/Kconfig | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/smsc/Kconfig b/drivers/net/ethernet/smsc/Kconfig
index 63aca9f847e1..4c2f612e4414 100644
--- a/drivers/net/ethernet/smsc/Kconfig
+++ b/drivers/net/ethernet/smsc/Kconfig
@@ -20,7 +20,7 @@ if NET_VENDOR_SMSC
config SMC9194
tristate "SMC 9194 support"
- depends on (ISA || MAC && BROKEN)
+ depends on ISA
select CRC32
---help---
This is support for the SMC9xxx based Ethernet cards. Choose this
--
2.13.5
^ permalink raw reply related
* [PATCH RESEND net 0/9] Fixes, cleanup and modernization for some legacy ethernet NIC drivers
From: Finn Thain @ 2017-10-03 1:07 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, linux-kernel
This patch series fixes some logging bugs and adds some missing message
severity levels.
There are also cleanup patches for dead code and some Kconfig cruft.
Custom debug message logging is converted to netif_* calls to reduce
code duplication.
All up, about 150 lines of code are eliminated.
My apologies for duplicated messages. I messed up the addressing.
Finn Thain (9):
net/smc9194: Remove bogus CONFIG_MAC reference
net/mac89x0: Remove dead or unreachable code
net/mac89x0: Fix and modernize log messages
net/mac89x0: Replace custom debug logging with netif_* calls
net/macmace: Fix and cleanup log messages
net/sonic: Cleanup and modernize log messages
net/sonic: Replace custom debug logging with netif_* calls
net/8390: Fix redundant code
net/mac8390: Fix log messages
drivers/net/ethernet/8390/ax88796.c | 3 -
drivers/net/ethernet/8390/axnet_cs.c | 2 -
drivers/net/ethernet/8390/etherh.c | 17 -----
drivers/net/ethernet/8390/hydra.c | 4 --
drivers/net/ethernet/8390/lib8390.c | 2 +
drivers/net/ethernet/8390/mac8390.c | 42 +++++-------
drivers/net/ethernet/8390/mcf8390.c | 4 --
drivers/net/ethernet/8390/pcnet_cs.c | 4 --
drivers/net/ethernet/8390/zorro8390.c | 5 --
drivers/net/ethernet/apple/macmace.c | 10 +--
drivers/net/ethernet/cirrus/mac89x0.c | 109 ++++++++-----------------------
drivers/net/ethernet/natsemi/jazzsonic.c | 26 ++------
drivers/net/ethernet/natsemi/macsonic.c | 45 ++++---------
drivers/net/ethernet/natsemi/sonic.c | 92 +++++++++++++-------------
drivers/net/ethernet/natsemi/sonic.h | 2 +
drivers/net/ethernet/natsemi/xtsonic.c | 28 ++------
drivers/net/ethernet/smsc/Kconfig | 2 +-
17 files changed, 123 insertions(+), 274 deletions(-)
--
2.13.5
^ permalink raw reply
* Re: [net-next V3 PATCH 3/5] bpf: cpumap xdp_buff to skb conversion and allocation
From: Alexei Starovoitov @ 2017-10-03 1:02 UTC (permalink / raw)
To: Jesper Dangaard Brouer
Cc: netdev, jakub.kicinski, Michael S. Tsirkin, pavel.odintsov,
Jason Wang, mchan, John Fastabend, peter.waskiewicz.jr,
Daniel Borkmann, Andy Gospodarek
In-Reply-To: <150696032966.24152.3096148163967111058.stgit@firesoul>
On Mon, Oct 02, 2017 at 06:05:29PM +0200, Jesper Dangaard Brouer wrote:
> + while ((xdp_pkt = __ptr_ring_consume(rcpu->queue))) {
> + struct sk_buff *skb;
> + int ret;
> +
> + /* Allow busy polling again */
> + empty_cnt = 0;
> +
> + skb = cpu_map_build_skb(rcpu, xdp_pkt);
> + if (!skb) {
> + page_frag_free(xdp_pkt);
> + continue;
> + }
> +
> + /* Inject into network stack */
> + ret = netif_receive_skb(skb);
> + if (ret == NET_RX_DROP)
> + drops++;
I thought the whole thing is an alternative to RPS,
but netif_receive_skb_internal() will call into RPS logic.
So the user has to make sure it disabled or they will
conflict in some weird way?
Or you're calling netif_receive_skb() to be able to call
generic XDP on that cpu again ?
But that prog can do cpumap redirect again?
sort-of recursive redirect? Is it really useful?
May be call into __netif_receive_skb_core() directly?
not sure.
I'm asking all these questions to make sure we think through
these implications before it becomes an abi.
^ permalink raw reply
* [PATCH net] net: br: Fix igmp snooping offload with CONFIG_BRIDGE_VLAN_FILTERING
From: Andrew Lunn @ 2017-10-03 0:55 UTC (permalink / raw)
To: David Miller; +Cc: Vivien Didelot, netdev, Andrew Lunn
With CONFIG_BRIDGE_VLAN_FILTERING enabled, but the feature not enabled
via /sys/class/net/brX/bridge/vlan_filtering, mdb offloaded to the
kernel have the wrong VID.
When an interface is added to the bridge, switchdev is first used to
notify the hardware that a port has joined a bridge. This is
immediately followed by the default_pvid, 1, being added to the
interface via another switchdev call.
The bridge will then perform IGMP snooping, and offload an mdb entries
to the switch as needed. With vlan filtering disabled, the vid is left
as 0. This causes the switch to put the static mdb into the wrong
vlan, and so frames are not forwarded by the mdb entry.
If vlan filtering is disable, use the default_pvid, not 0.
Fixes: f1fecb1d10ec ("bridge: Reflect MDB entries to hardware")
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
---
net/bridge/br_vlan.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/net/bridge/br_vlan.c b/net/bridge/br_vlan.c
index 233a30040c91..aa3589891797 100644
--- a/net/bridge/br_vlan.c
+++ b/net/bridge/br_vlan.c
@@ -492,6 +492,7 @@ bool br_allowed_ingress(const struct net_bridge *br,
*/
if (!br->vlan_enabled) {
BR_INPUT_SKB_CB(skb)->vlan_filtered = false;
+ *vid = br_get_pvid(vg);
return true;
}
--
2.14.1
^ permalink raw reply related
* Re: [PATCH net-next v2 3/3] tools: bpftool: add documentation
From: Alexei Starovoitov @ 2017-10-03 0:55 UTC (permalink / raw)
To: Jakub Kicinski; +Cc: netdev, daniel, oss-drivers, David Beckett
In-Reply-To: <20171002231130.12406-4-jakub.kicinski@netronome.com>
On Mon, Oct 02, 2017 at 04:11:30PM -0700, Jakub Kicinski wrote:
> Add documentation for bpftool. Separate files for each subcommand.
> Use rst format. Documentation is compiled into man pages using
> rst2man.
>
> Signed-off-by: David Beckett <david.beckett@netronome.com>
> Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
> ---
> tools/bpf/bpftool/Documentation/Makefile | 34 +++++++
> tools/bpf/bpftool/Documentation/bpftool-map.txt | 110 +++++++++++++++++++++++
> tools/bpf/bpftool/Documentation/bpftool-prog.txt | 81 +++++++++++++++++
> tools/bpf/bpftool/Documentation/bpftool.txt | 34 +++++++
> tools/bpf/bpftool/Makefile | 6 ++
> 5 files changed, 265 insertions(+)
> create mode 100644 tools/bpf/bpftool/Documentation/Makefile
> create mode 100644 tools/bpf/bpftool/Documentation/bpftool-map.txt
> create mode 100644 tools/bpf/bpftool/Documentation/bpftool-prog.txt
> create mode 100644 tools/bpf/bpftool/Documentation/bpftool.txt
>
> diff --git a/tools/bpf/bpftool/Documentation/Makefile b/tools/bpf/bpftool/Documentation/Makefile
> new file mode 100644
> index 000000000000..ebd21ab2c6df
> --- /dev/null
> +++ b/tools/bpf/bpftool/Documentation/Makefile
> @@ -0,0 +1,34 @@
> +include ../../../scripts/Makefile.include
> +include ../../../scripts/utilities.mak
> +
> +INSTALL ?= install
> +RM ?= rm -f
> +
> +# Make the path relative to DESTDIR, not prefix
> +ifndef DESTDIR
> +prefix?=$(HOME)
> +endif
> +mandir ?= $(prefix)/share/man
> +man8dir = $(mandir)/man8
> +
> +MAN8_TXT = $(wildcard *.txt)
> +
> +_DOC_MAN8 = $(patsubst %.txt,%.8,$(MAN8_TXT))
> +DOC_MAN8 = $(addprefix $(OUTPUT),$(_DOC_MAN8))
> +
> +man: man8
> +man8: $(DOC_MAN8)
> +
> +$(OUTPUT)%.8: %.txt
> + rst2man $< > $@
> +
> +clean:
> + $(call QUIET_CLEAN, Documentation) $(RM) $(DOC_MAN8)
> +
> +install: man
> + $(call QUIET_INSTALL, Documentation-man) \
> + $(INSTALL) -d -m 755 $(DESTDIR)$(man8dir); \
> + $(INSTALL) -m 644 $(DOC_MAN8) $(DESTDIR)$(man8dir);
> +
> +.PHONY: man man8 clean install
> +.DEFAULT_GOAL := man
> diff --git a/tools/bpf/bpftool/Documentation/bpftool-map.txt b/tools/bpf/bpftool/Documentation/bpftool-map.txt
> new file mode 100644
> index 000000000000..ea82b9059a0a
> --- /dev/null
> +++ b/tools/bpf/bpftool/Documentation/bpftool-map.txt
> @@ -0,0 +1,110 @@
> +================
> +bpftool-map
> +================
> +-------------------------------------------------------------------------------
> +tool for inspection and simple manipulation of eBPF maps
> +-------------------------------------------------------------------------------
> +
> +:Manual section: 8
> +
> +SYNOPSIS
> +========
> +
> + **bpftool** **map** *COMMAND*
> +
> + *COMMANDS* :=
> + { show | dump | update | lookup | getnext | delete | pin | help }
> +
> +MAP COMMANDS
> +=============
> +
> +| **bpftool** map show [*MAP*]
> +| **bpftool** map dump *MAP*
> +| **bpftool** map update *MAP* key *BYTES* value *VALUE* [*UPDATE_FLAGS*]
> +| **bpftool** map lookup *MAP* key *BYTES*
> +| **bpftool** map getnext *MAP* [key *BYTES*]
> +| **bpftool** map delete *MAP* key *BYTES*
> +| **bpftool** map pin *MAP* *FILE*
> +| **bpftool** map help
> +|
> +| *MAP* := { id MAP_ID | pinned FILE }
> +| *VALUE* := { BYTES | MAP | PROGRAM }
> +| *UPDATE_FLAGS* := { any | exist | noexist }
> +
> +DESCRIPTION
> +===========
> + **bpftool map show** [*MAP*]
> + Show information about loaded maps. If *MAP* is specified
> + show information only about given map, otherwise list all
> + maps currently loaded on the system.
> +
> + Output will start with map ID followed by map type and
> + zero or more named attributes (depending on kernel version).
> +
> + **bpftool map dump** *MAP*
> + Dump all entries in a given *MAP*.
> +
> + **bpftool map update** *MAP* **key** *BYTES* **value** *VALUE* [*UPDATE_FLAGS*]
> + Update map entry for a given *KEY*.
> +
> + *UPDATE_FLAGS* can be one of: **any** update existing entry
> + or add if doesn't exit; **exist** update only if entry already
> + exists; **noexist** update only if entry doesn't exist.
> +
> + **bpftool map lookup** *MAP* **key** *BYTES*
> + Lookup **key** in the map.
> +
> + **bpftool map getnext** *MAP* [**key** *BYTES*]
> + Get next key. If *key* is not specified, get first key.
> +
> + **bpftool map delete** *MAP* **key** *BYTES*
> + Remove entry from the map.
> +
> + **bpftool map pin** *MAP* *FILE*
> + Pin map *MAP* as *FILE*.
> +
> + Note: *FILE* must be located in *bpffs* mount.
> +
> + **bpftool map help**
> + Print short help message.
> +
> +EXAMPLES
> +========
> +**# bpftool map show**
> +::
> +
> + 10: hash name:some_map flags:0x0
> + key:4B value:8B max_entries:2048 memlock:167936B
> +
> +**# bpftool map update id 10 key 13 00 07 00 value 02 00 00 00 01 02 03 04**
> +
> +**# bpftool map lookup id 10 key 0 1 2 3**
> +
> +::
> +
> + key: 00 01 02 03 value: 00 01 02 03 04 05 06 07
> +
> +
> +**# bpftool map dump id 10**
> +::
> +
> + key: 00 01 02 03 value: 00 01 02 03 04 05 06 07
> + key: 0d 00 07 00 value: 02 00 00 00 01 02 03 04
> + Found 2 elements
> +
> +**# bpftool map getnext id 10 key 0 1 2 3**
> +::
> +
> + key:
> + 00 01 02 03
> + next key:
> + 0d 00 07 00
> +
> +|
> +| **# mount -t bpf none /sys/fs/bpf/**
> +| **# bpftool map pin id 10 /sys/fs/bpf/map**
> +| **# bpftool map del pinned /sys/fs/bpf/map key 13 00 07 00**
> +
> +SEE ALSO
> +========
> + **bpftool**\ (8), **bpftool-prog**\ (8)
> diff --git a/tools/bpf/bpftool/Documentation/bpftool-prog.txt b/tools/bpf/bpftool/Documentation/bpftool-prog.txt
> new file mode 100644
> index 000000000000..d632de0e0212
> --- /dev/null
> +++ b/tools/bpf/bpftool/Documentation/bpftool-prog.txt
> @@ -0,0 +1,81 @@
> +================
> +bpftool-prog
> +================
> +-------------------------------------------------------------------------------
> +tool for inspection and simple manipulation of eBPF progs
> +-------------------------------------------------------------------------------
> +
> +:Manual section: 8
> +
> +SYNOPSIS
> +========
> +
> +| **bpftool** prog show [*PROG*]
> +| **bpftool** prog dump xlated *PROG* file *FILE*
> +| **bpftool** prog dump jited *PROG* [file *FILE*] [opcodes]
> +| **bpftool** prog pin *PROG* *FILE*
> +| **bpftool** prog help
> +|
> +| *PROG* := { id *PROG_ID* | pinned *FILE* | tag *PROG_TAG* }
> +
> +DESCRIPTION
> +===========
> + **bpftool prog show** [*PROG*]
> + Show information about loaded programs. If *PROG* is
> + specified show information only about given program, otherwise
> + list all programs currently loaded on the system.
> +
> + Output will start with program ID followed by program type and
> + zero or more named attributes (depending on kernel version).
> +
> + **bpftool prog dump xlated** *PROG* **file** *FILE*
> + Dump eBPF instructions of the program from the kernel to a
> + file.
> +
> + **bpftool prog dump jited** *PROG* [**file** *FILE*] [**opcodes**]
> + Dump jited image (host machine code) of the program.
> + If *FILE* is specified image will be written to a file,
> + otherwise it will be disassembled and printed to stdout.
> +
> + **opcodes** controls if raw opcodes will be printed.
> +
> + **bpftool prog pin** *PROG* *FILE*
> + Pin program *PROG* as *FILE*.
> +
> + Note: *FILE* must be located in *bpffs* mount.
> +
> + **bpftool prog help**
> + Print short help message.
> +
> +EXAMPLES
> +========
> +**# bpftool prog show**
> +::
> +
> + 10: xdp name:some_prog tag 00:5a:3d:21:23:62:0c:8b
could you please remove ':' in the output to match what
show_fdinfo and kallsyms do ?
> + loaded_at:2024.771 uid:0
may be translate that to something human readable?
> + xlated:528B jited:370B memlock:4096B map_ids:10
> +
> +|
> +| **# bpftool prog dump xlated id 10 file /tmp/t**
> +| **# ls -l /tmp/t**
> +| -rw------- 1 root root 560 Jul 22 01:42 /tmp/t
> +
> +|
> +| **# mount -t bpf none /sys/fs/bpf/**
> +| **# bpftool prog pin id 10 /sys/fs/bpf/prog**
> +| **# bpftool prog dum jited pinned /sys/fs/bpf/prog**
> +
> +::
> +
> + push %rbp
> + mov %rsp,%rbp
> + sub $0x228,%rsp
> + sub $0x28,%rbp
> + mov %rbx,0x0(%rbp)
imo too many steps to dump disasm output.
Can it print it if we just say:
bpftool prog dump jited id 10
and
dump xlated
will pretty print them as verifier output as well?
All that can be changed later. Thanks for the doc.
Acked-by: Alexei Starovoitov <ast@kernel.org>
^ permalink raw reply
* Re: [PATCH net v1 1/2] ARM: dts: imx: let's name the ptp interrupt for the fec ethernet driver
From: Andrew Lunn @ 2017-10-03 0:51 UTC (permalink / raw)
To: Troy Kisky; +Cc: shawn.guo, fugang.duan, netdev, davem, fabio.estevam, lznuaa
In-Reply-To: <20171003000442.31841-1-troy.kisky@boundarydevices.com>
On Mon, Oct 02, 2017 at 05:04:41PM -0700, Troy Kisky wrote:
> imx7s/imx7d has the ptp interrupt newly added as well.
> This will allow the ptp interrupt to have its own interrupt routine.
>
> Signed-off-by: Troy Kisky <troy.kisky@boundarydevices.com>
> ---
> arch/arm/boot/dts/imx6qdl.dtsi | 1 +
> arch/arm/boot/dts/imx6sx.dtsi | 2 ++
> arch/arm/boot/dts/imx6ul.dtsi | 2 ++
> arch/arm/boot/dts/imx7d.dtsi | 4 +++-
> arch/arm/boot/dts/imx7s.dtsi | 4 +++-
> 5 files changed, 11 insertions(+), 2 deletions(-)
>
> diff --git a/arch/arm/boot/dts/imx6qdl.dtsi b/arch/arm/boot/dts/imx6qdl.dtsi
> index 8884b4a3cafb..d848d2bfe8e2 100644
> --- a/arch/arm/boot/dts/imx6qdl.dtsi
> +++ b/arch/arm/boot/dts/imx6qdl.dtsi
> @@ -1017,6 +1017,7 @@
> fec: ethernet@02188000 {
> compatible = "fsl,imx6q-fec";
> reg = <0x02188000 0x4000>;
> + interrupt-names = "","ptp";
Hi Troy
The "" looks a bit odd. Can you use a name here?
Andrew
^ permalink raw reply
* [PATCH net v1 2/2] net: fec: Let fec_ptp have its own interrupt routine
From: Troy Kisky @ 2017-10-03 0:04 UTC (permalink / raw)
To: shawn.guo, fugang.duan, netdev, davem; +Cc: fabio.estevam, lznuaa, Troy Kisky
In-Reply-To: <20171003000442.31841-1-troy.kisky@boundarydevices.com>
This is better for code locality and should slightly
speed up normal interrupts.
This also allows PPS clock output to start working for
i.mx7. This is because i.mx7 was already using the limit
of 3 interrupts, and needed another.
Signed-off-by: Troy Kisky <troy.kisky@boundarydevices.com>
---
If this patch is taken before the corresponding dtb file patch
than time stamping will be momentarily broken.
---
drivers/net/ethernet/freescale/fec.h | 1 -
drivers/net/ethernet/freescale/fec_main.c | 15 +++---
drivers/net/ethernet/freescale/fec_ptp.c | 77 ++++++++++++++++++-------------
3 files changed, 52 insertions(+), 41 deletions(-)
diff --git a/drivers/net/ethernet/freescale/fec.h b/drivers/net/ethernet/freescale/fec.h
index ede1876a9a19..782509041102 100644
--- a/drivers/net/ethernet/freescale/fec.h
+++ b/drivers/net/ethernet/freescale/fec.h
@@ -587,7 +587,6 @@ void fec_ptp_stop(struct platform_device *pdev);
void fec_ptp_start_cyclecounter(struct net_device *ndev);
int fec_ptp_set(struct net_device *ndev, struct ifreq *ifr);
int fec_ptp_get(struct net_device *ndev, struct ifreq *ifr);
-uint fec_ptp_check_pps_event(struct fec_enet_private *fep);
/****************************************************************************/
#endif /* FEC_H */
diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c
index 3dc2d771a222..80fe04165ba0 100644
--- a/drivers/net/ethernet/freescale/fec_main.c
+++ b/drivers/net/ethernet/freescale/fec_main.c
@@ -1602,10 +1602,6 @@ fec_enet_interrupt(int irq, void *dev_id)
ret = IRQ_HANDLED;
complete(&fep->mdio_done);
}
-
- if (fep->ptp_clock)
- if (fec_ptp_check_pps_event(fep))
- ret = IRQ_HANDLED;
return ret;
}
@@ -3325,6 +3321,7 @@ fec_probe(struct platform_device *pdev)
struct device_node *np = pdev->dev.of_node, *phy_node;
int num_tx_qs;
int num_rx_qs;
+ int ptp_irq, j;
fec_enet_get_queue_num(pdev, &num_tx_qs, &num_rx_qs);
@@ -3472,20 +3469,24 @@ fec_probe(struct platform_device *pdev)
if (ret)
goto failed_init;
- for (i = 0; i < FEC_IRQ_NUM; i++) {
- irq = platform_get_irq(pdev, i);
+ ptp_irq = platform_get_irq_byname(pdev, "ptp");
+ i = j = 0;
+ while (i < FEC_IRQ_NUM) {
+ irq = platform_get_irq(pdev, j++);
if (irq < 0) {
if (i)
break;
ret = irq;
goto failed_irq;
}
+ if (irq == ptp_irq)
+ continue;
ret = devm_request_irq(&pdev->dev, irq, fec_enet_interrupt,
0, pdev->name, ndev);
if (ret)
goto failed_irq;
- fep->irq[i] = irq;
+ fep->irq[i++] = irq;
}
init_completion(&fep->mdio_done);
diff --git a/drivers/net/ethernet/freescale/fec_ptp.c b/drivers/net/ethernet/freescale/fec_ptp.c
index 6ebad3fac81d..20e01a3ca453 100644
--- a/drivers/net/ethernet/freescale/fec_ptp.c
+++ b/drivers/net/ethernet/freescale/fec_ptp.c
@@ -549,6 +549,38 @@ static void fec_time_keep(struct work_struct *work)
schedule_delayed_work(&fep->time_keep, HZ);
}
+/*
+ * This function checks the pps event and reloads the timer compare counter.
+ */
+static irqreturn_t fec_ptp_interrupt(int irq, void *dev_id)
+{
+ struct net_device *ndev = dev_id;
+ struct fec_enet_private *fep = netdev_priv(ndev);
+ u32 val;
+ u8 channel = fep->pps_channel;
+ struct ptp_clock_event event;
+
+ val = readl(fep->hwp + FEC_TCSR(channel));
+ if (val & FEC_T_TF_MASK) {
+ /* Write the next next compare(not the next according the spec)
+ * value to the register
+ */
+ writel(fep->next_counter, fep->hwp + FEC_TCCR(channel));
+ do {
+ writel(val, fep->hwp + FEC_TCSR(channel));
+ } while (readl(fep->hwp + FEC_TCSR(channel)) & FEC_T_TF_MASK);
+
+ /* Update the counter; */
+ fep->next_counter = (fep->next_counter + fep->reload_period) & fep->cc.mask;
+
+ event.type = PTP_CLOCK_PPS;
+ ptp_clock_event(fep->ptp_clock, &event);
+ return IRQ_HANDLED;
+ }
+
+ return IRQ_NONE;
+}
+
/**
* fec_ptp_init
* @ndev: The FEC network adapter
@@ -562,6 +594,8 @@ void fec_ptp_init(struct platform_device *pdev)
{
struct net_device *ndev = platform_get_drvdata(pdev);
struct fec_enet_private *fep = netdev_priv(ndev);
+ int irq;
+ int ret;
fep->ptp_caps.owner = THIS_MODULE;
snprintf(fep->ptp_caps.name, 16, "fec ptp");
@@ -587,6 +621,16 @@ void fec_ptp_init(struct platform_device *pdev)
INIT_DELAYED_WORK(&fep->time_keep, fec_time_keep);
+ irq = platform_get_irq_byname(pdev, "ptp");
+ if (irq < 0)
+ return;
+ ret = devm_request_irq(&pdev->dev, irq, fec_ptp_interrupt,
+ 0, pdev->name, ndev);
+ if (ret < 0) {
+ dev_warn(&pdev->dev, "request for ptp irq failed(%d)\n", ret);
+ return;
+ }
+
fep->ptp_clock = ptp_clock_register(&fep->ptp_caps, &pdev->dev);
if (IS_ERR(fep->ptp_clock)) {
fep->ptp_clock = NULL;
@@ -605,36 +649,3 @@ void fec_ptp_stop(struct platform_device *pdev)
if (fep->ptp_clock)
ptp_clock_unregister(fep->ptp_clock);
}
-
-/**
- * fec_ptp_check_pps_event
- * @fep: the fec_enet_private structure handle
- *
- * This function check the pps event and reload the timer compare counter.
- */
-uint fec_ptp_check_pps_event(struct fec_enet_private *fep)
-{
- u32 val;
- u8 channel = fep->pps_channel;
- struct ptp_clock_event event;
-
- val = readl(fep->hwp + FEC_TCSR(channel));
- if (val & FEC_T_TF_MASK) {
- /* Write the next next compare(not the next according the spec)
- * value to the register
- */
- writel(fep->next_counter, fep->hwp + FEC_TCCR(channel));
- do {
- writel(val, fep->hwp + FEC_TCSR(channel));
- } while (readl(fep->hwp + FEC_TCSR(channel)) & FEC_T_TF_MASK);
-
- /* Update the counter; */
- fep->next_counter = (fep->next_counter + fep->reload_period) & fep->cc.mask;
-
- event.type = PTP_CLOCK_PPS;
- ptp_clock_event(fep->ptp_clock, &event);
- return 1;
- }
-
- return 0;
-}
--
2.11.0
^ permalink raw reply related
* [PATCH net v1 1/2] ARM: dts: imx: let's name the ptp interrupt for the fec ethernet driver
From: Troy Kisky @ 2017-10-03 0:04 UTC (permalink / raw)
To: shawn.guo, fugang.duan, netdev, davem; +Cc: fabio.estevam, lznuaa, Troy Kisky
imx7s/imx7d has the ptp interrupt newly added as well.
This will allow the ptp interrupt to have its own interrupt routine.
Signed-off-by: Troy Kisky <troy.kisky@boundarydevices.com>
---
arch/arm/boot/dts/imx6qdl.dtsi | 1 +
arch/arm/boot/dts/imx6sx.dtsi | 2 ++
arch/arm/boot/dts/imx6ul.dtsi | 2 ++
arch/arm/boot/dts/imx7d.dtsi | 4 +++-
arch/arm/boot/dts/imx7s.dtsi | 4 +++-
5 files changed, 11 insertions(+), 2 deletions(-)
diff --git a/arch/arm/boot/dts/imx6qdl.dtsi b/arch/arm/boot/dts/imx6qdl.dtsi
index 8884b4a3cafb..d848d2bfe8e2 100644
--- a/arch/arm/boot/dts/imx6qdl.dtsi
+++ b/arch/arm/boot/dts/imx6qdl.dtsi
@@ -1017,6 +1017,7 @@
fec: ethernet@02188000 {
compatible = "fsl,imx6q-fec";
reg = <0x02188000 0x4000>;
+ interrupt-names = "","ptp";
interrupts-extended =
<&intc 0 118 IRQ_TYPE_LEVEL_HIGH>,
<&intc 0 119 IRQ_TYPE_LEVEL_HIGH>;
diff --git a/arch/arm/boot/dts/imx6sx.dtsi b/arch/arm/boot/dts/imx6sx.dtsi
index 6c7eb54be9e2..d64ae33c0b41 100644
--- a/arch/arm/boot/dts/imx6sx.dtsi
+++ b/arch/arm/boot/dts/imx6sx.dtsi
@@ -861,6 +861,7 @@
fec1: ethernet@02188000 {
compatible = "fsl,imx6sx-fec", "fsl,imx6q-fec";
reg = <0x02188000 0x4000>;
+ interrupt-names = "","ptp";
interrupts = <GIC_SPI 118 IRQ_TYPE_LEVEL_HIGH>,
<GIC_SPI 119 IRQ_TYPE_LEVEL_HIGH>;
clocks = <&clks IMX6SX_CLK_ENET>,
@@ -970,6 +971,7 @@
fec2: ethernet@021b4000 {
compatible = "fsl,imx6sx-fec", "fsl,imx6q-fec";
reg = <0x021b4000 0x4000>;
+ interrupt-names = "","ptp";
interrupts = <GIC_SPI 102 IRQ_TYPE_LEVEL_HIGH>,
<GIC_SPI 103 IRQ_TYPE_LEVEL_HIGH>;
clocks = <&clks IMX6SX_CLK_ENET>,
diff --git a/arch/arm/boot/dts/imx6ul.dtsi b/arch/arm/boot/dts/imx6ul.dtsi
index f11a241a340d..fe093f8531b0 100644
--- a/arch/arm/boot/dts/imx6ul.dtsi
+++ b/arch/arm/boot/dts/imx6ul.dtsi
@@ -476,6 +476,7 @@
fec2: ethernet@020b4000 {
compatible = "fsl,imx6ul-fec", "fsl,imx6q-fec";
reg = <0x020b4000 0x4000>;
+ interrupt-names = "","ptp";
interrupts = <GIC_SPI 120 IRQ_TYPE_LEVEL_HIGH>,
<GIC_SPI 121 IRQ_TYPE_LEVEL_HIGH>;
clocks = <&clks IMX6UL_CLK_ENET>,
@@ -775,6 +776,7 @@
fec1: ethernet@02188000 {
compatible = "fsl,imx6ul-fec", "fsl,imx6q-fec";
reg = <0x02188000 0x4000>;
+ interrupt-names = "","ptp";
interrupts = <GIC_SPI 118 IRQ_TYPE_LEVEL_HIGH>,
<GIC_SPI 119 IRQ_TYPE_LEVEL_HIGH>;
clocks = <&clks IMX6UL_CLK_ENET>,
diff --git a/arch/arm/boot/dts/imx7d.dtsi b/arch/arm/boot/dts/imx7d.dtsi
index f46814a7ea44..e8fc76483236 100644
--- a/arch/arm/boot/dts/imx7d.dtsi
+++ b/arch/arm/boot/dts/imx7d.dtsi
@@ -114,9 +114,11 @@
fec2: ethernet@30bf0000 {
compatible = "fsl,imx7d-fec", "fsl,imx6sx-fec";
reg = <0x30bf0000 0x10000>;
+ interrupt-names = "","","","ptp";
interrupts = <GIC_SPI 100 IRQ_TYPE_LEVEL_HIGH>,
<GIC_SPI 101 IRQ_TYPE_LEVEL_HIGH>,
- <GIC_SPI 102 IRQ_TYPE_LEVEL_HIGH>;
+ <GIC_SPI 102 IRQ_TYPE_LEVEL_HIGH>,
+ <GIC_SPI 103 IRQ_TYPE_LEVEL_HIGH>;
clocks = <&clks IMX7D_ENET_AXI_ROOT_CLK>,
<&clks IMX7D_ENET_AXI_ROOT_CLK>,
<&clks IMX7D_ENET2_TIME_ROOT_CLK>,
diff --git a/arch/arm/boot/dts/imx7s.dtsi b/arch/arm/boot/dts/imx7s.dtsi
index 82ad26e766eb..fc1376079494 100644
--- a/arch/arm/boot/dts/imx7s.dtsi
+++ b/arch/arm/boot/dts/imx7s.dtsi
@@ -1007,9 +1007,11 @@
fec1: ethernet@30be0000 {
compatible = "fsl,imx7d-fec", "fsl,imx6sx-fec";
reg = <0x30be0000 0x10000>;
+ interrupt-names = "","","","ptp";
interrupts = <GIC_SPI 118 IRQ_TYPE_LEVEL_HIGH>,
<GIC_SPI 119 IRQ_TYPE_LEVEL_HIGH>,
- <GIC_SPI 120 IRQ_TYPE_LEVEL_HIGH>;
+ <GIC_SPI 120 IRQ_TYPE_LEVEL_HIGH>,
+ <GIC_SPI 121 IRQ_TYPE_LEVEL_HIGH>;
clocks = <&clks IMX7D_ENET_AXI_ROOT_CLK>,
<&clks IMX7D_ENET_AXI_ROOT_CLK>,
<&clks IMX7D_ENET1_TIME_ROOT_CLK>,
--
2.11.0
^ permalink raw reply related
* [PATCH net-next 6/8] libbpf: sync bpf.h
From: Alexei Starovoitov @ 2017-10-02 23:48 UTC (permalink / raw)
To: David S . Miller
Cc: Daniel Borkmann, Tejun Heo, David Ahern, netdev, kernel-team
In-Reply-To: <20171002234857.3707580-1-ast@fb.com>
tools/include/uapi/linux/bpf.h got out of sync with actual kernel header.
Update it.
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
---
tools/include/uapi/linux/bpf.h | 55 +++++++++++++++++++++++++++++++++++++++---
1 file changed, 52 insertions(+), 3 deletions(-)
diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h
index 6d2137b4cf38..cb2b9f95160a 100644
--- a/tools/include/uapi/linux/bpf.h
+++ b/tools/include/uapi/linux/bpf.h
@@ -92,6 +92,7 @@ enum bpf_cmd {
BPF_PROG_GET_FD_BY_ID,
BPF_MAP_GET_FD_BY_ID,
BPF_OBJ_GET_INFO_BY_FD,
+ BPF_PROG_QUERY,
};
enum bpf_map_type {
@@ -143,11 +144,47 @@ enum bpf_attach_type {
#define MAX_BPF_ATTACH_TYPE __MAX_BPF_ATTACH_TYPE
-/* If BPF_F_ALLOW_OVERRIDE flag is used in BPF_PROG_ATTACH command
- * to the given target_fd cgroup the descendent cgroup will be able to
- * override effective bpf program that was inherited from this cgroup
+/* cgroup-bpf attach flags used in BPF_PROG_ATTACH command
+ *
+ * NONE(default): No further bpf programs allowed in the subtree.
+ *
+ * BPF_F_ALLOW_OVERRIDE: If a sub-cgroup installs some bpf program,
+ * the program in this cgroup yields to sub-cgroup program.
+ *
+ * BPF_F_ALLOW_MULTI: If a sub-cgroup installs some bpf program,
+ * that cgroup program gets run in addition to the program in this cgroup.
+ *
+ * Only one program is allowed to be attached to a cgroup with
+ * NONE or BPF_F_ALLOW_OVERRIDE flag.
+ * Attaching another program on top of NONE or BPF_F_ALLOW_OVERRIDE will
+ * release old program and attach the new one. Attach flags has to match.
+ *
+ * Multiple programs are allowed to be attached to a cgroup with
+ * BPF_F_ALLOW_MULTI flag. They are executed in FIFO order
+ * (those that were attached first, run first)
+ * The programs of sub-cgroup are executed first, then programs of
+ * this cgroup and then programs of parent cgroup.
+ * When children program makes decision (like picking TCP CA or sock bind)
+ * parent program has a chance to override it.
+ *
+ * A cgroup with MULTI or OVERRIDE flag allows any attach flags in sub-cgroups.
+ * A cgroup with NONE doesn't allow any programs in sub-cgroups.
+ * Ex1:
+ * cgrp1 (MULTI progs A, B) ->
+ * cgrp2 (OVERRIDE prog C) ->
+ * cgrp3 (MULTI prog D) ->
+ * cgrp4 (OVERRIDE prog E) ->
+ * cgrp5 (NONE prog F)
+ * the event in cgrp5 triggers execution of F,D,A,B in that order.
+ * if prog F is detached, the execution is E,D,A,B
+ * if prog F and D are detached, the execution is E,A,B
+ * if prog F, E and D are detached, the execution is C,A,B
+ *
+ * All eligible programs are executed regardless of return code from
+ * earlier programs.
*/
#define BPF_F_ALLOW_OVERRIDE (1U << 0)
+#define BPF_F_ALLOW_MULTI (1U << 1)
/* If BPF_F_STRICT_ALIGNMENT is used in BPF_PROG_LOAD command, the
* verifier will perform strict alignment checking as if the kernel
@@ -175,6 +212,9 @@ enum bpf_attach_type {
/* Specify numa node during map creation */
#define BPF_F_NUMA_NODE (1U << 2)
+/* flags for BPF_PROG_QUERY */
+#define BPF_F_QUERY_EFFECTIVE (1U << 0)
+
#define BPF_OBJ_NAME_LEN 16U
union bpf_attr {
@@ -253,6 +293,15 @@ union bpf_attr {
__u32 info_len;
__aligned_u64 info;
} info;
+
+ struct { /* anonymous struct used by BPF_PROG_QUERY command */
+ __u32 target_fd; /* container object to query */
+ __u32 attach_type;
+ __u32 query_flags;
+ __u32 attach_flags;
+ __aligned_u64 prog_ids;
+ __u32 prog_cnt;
+ } query;
} __attribute__((aligned(8)));
/* BPF helper function descriptions:
--
2.9.5
^ permalink raw reply related
* [PATCH net-next 7/8] libbpf: add support for BPF_PROG_QUERY
From: Alexei Starovoitov @ 2017-10-02 23:48 UTC (permalink / raw)
To: David S . Miller
Cc: Daniel Borkmann, Tejun Heo, David Ahern, netdev, kernel-team
In-Reply-To: <20171002234857.3707580-1-ast@fb.com>
add support for BPF_PROG_QUERY command to libbpf
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
---
tools/lib/bpf/bpf.c | 20 ++++++++++++++++++++
tools/lib/bpf/bpf.h | 3 ++-
2 files changed, 22 insertions(+), 1 deletion(-)
diff --git a/tools/lib/bpf/bpf.c b/tools/lib/bpf/bpf.c
index d4b6ba8292ee..5128677e4117 100644
--- a/tools/lib/bpf/bpf.c
+++ b/tools/lib/bpf/bpf.c
@@ -303,6 +303,26 @@ int bpf_prog_detach2(int prog_fd, int target_fd, enum bpf_attach_type type)
return sys_bpf(BPF_PROG_DETACH, &attr, sizeof(attr));
}
+int bpf_prog_query(int target_fd, enum bpf_attach_type type, __u32 query_flags,
+ __u32 *attach_flags, __u32 *prog_ids, __u32 *prog_cnt)
+{
+ union bpf_attr attr;
+ int ret;
+
+ bzero(&attr, sizeof(attr));
+ attr.query.target_fd = target_fd;
+ attr.query.attach_type = type;
+ attr.query.query_flags = query_flags;
+ attr.query.prog_cnt = *prog_cnt;
+ attr.query.prog_ids = ptr_to_u64(prog_ids);
+
+ ret = sys_bpf(BPF_PROG_QUERY, &attr, sizeof(attr));
+ if (attach_flags)
+ *attach_flags = attr.query.attach_flags;
+ *prog_cnt = attr.query.prog_cnt;
+ return ret;
+}
+
int bpf_prog_test_run(int prog_fd, int repeat, void *data, __u32 size,
void *data_out, __u32 *size_out, __u32 *retval,
__u32 *duration)
diff --git a/tools/lib/bpf/bpf.h b/tools/lib/bpf/bpf.h
index afd64727c9cf..6534889e2b2f 100644
--- a/tools/lib/bpf/bpf.h
+++ b/tools/lib/bpf/bpf.h
@@ -75,5 +75,6 @@ int bpf_map_get_next_id(__u32 start_id, __u32 *next_id);
int bpf_prog_get_fd_by_id(__u32 id);
int bpf_map_get_fd_by_id(__u32 id);
int bpf_obj_get_info_by_fd(int prog_fd, void *info, __u32 *info_len);
-
+int bpf_prog_query(int target_fd, enum bpf_attach_type type, __u32 query_flags,
+ __u32 *attach_flags, __u32 *prog_ids, __u32 *prog_cnt);
#endif
--
2.9.5
^ permalink raw reply related
* [PATCH net-next 5/8] samples/bpf: add multi-prog cgroup test case
From: Alexei Starovoitov @ 2017-10-02 23:48 UTC (permalink / raw)
To: David S . Miller
Cc: Daniel Borkmann, Tejun Heo, David Ahern, netdev, kernel-team
In-Reply-To: <20171002234857.3707580-1-ast@fb.com>
create 5 cgroups, attach 6 progs and check that progs are executed as:
cgrp1 (MULTI progs A, B) ->
cgrp2 (OVERRIDE prog C) ->
cgrp3 (MULTI prog D) ->
cgrp4 (OVERRIDE prog E) ->
cgrp5 (NONE prog F)
the event in cgrp5 triggers execution of F,D,A,B in that order.
if prog F is detached, the execution is E,D,A,B
if prog F and D are detached, the execution is E,A,B
if prog F, E and D are detached, the execution is C,A,B
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
---
samples/bpf/cgroup_helpers.c | 4 +-
samples/bpf/test_cgrp2_attach2.c | 188 +++++++++++++++++++++++++++++++++++++--
2 files changed, 185 insertions(+), 7 deletions(-)
diff --git a/samples/bpf/cgroup_helpers.c b/samples/bpf/cgroup_helpers.c
index 9d1be9426401..88bdcf4b1670 100644
--- a/samples/bpf/cgroup_helpers.c
+++ b/samples/bpf/cgroup_helpers.c
@@ -56,7 +56,7 @@ int setup_cgroup_environment(void)
return 1;
}
- if (mount("none", CGROUP_MOUNT_PATH, "cgroup2", 0, NULL)) {
+ if (mount("none", CGROUP_MOUNT_PATH, "cgroup2", 0, NULL) && errno != EBUSY) {
log_err("mount cgroup2");
return 1;
}
@@ -163,7 +163,7 @@ int create_and_get_cgroup(char *path)
format_cgroup_path(cgroup_path, path);
if (mkdir(cgroup_path, 0777) && errno != EEXIST) {
- log_err("mkdiring cgroup");
+ log_err("mkdiring cgroup %s .. %s", path, cgroup_path);
return 0;
}
diff --git a/samples/bpf/test_cgrp2_attach2.c b/samples/bpf/test_cgrp2_attach2.c
index 3049b1f26267..9a9f6836e5e9 100644
--- a/samples/bpf/test_cgrp2_attach2.c
+++ b/samples/bpf/test_cgrp2_attach2.c
@@ -30,7 +30,7 @@
#define FOO "/foo"
#define BAR "/foo/bar/"
-#define PING_CMD "ping -c1 -w1 127.0.0.1"
+#define PING_CMD "ping -c1 -w1 127.0.0.1 > /dev/null"
char bpf_log_buf[BPF_LOG_BUF_SIZE];
@@ -55,8 +55,7 @@ static int prog_load(int verdict)
return ret;
}
-
-int main(int argc, char **argv)
+static int test_foo_bar(void)
{
int drop_prog, allow_prog, foo = 0, bar = 0, rc = 0;
@@ -189,8 +188,187 @@ int main(int argc, char **argv)
close(bar);
cleanup_cgroup_environment();
if (!rc)
- printf("PASS\n");
+ printf("### override:PASS\n");
+ else
+ printf("### override:FAIL\n");
+ return rc;
+}
+
+static int map_fd = -1;
+
+static int prog_load_cnt(int verdict, int val)
+{
+ if (map_fd < 0)
+ map_fd = bpf_create_map(BPF_MAP_TYPE_ARRAY, 4, 8, 1, 0);
+ if (map_fd < 0) {
+ printf("failed to create map '%s'\n", strerror(errno));
+ return -1;
+ }
+
+ struct bpf_insn prog[] = {
+ BPF_MOV32_IMM(BPF_REG_0, 0),
+ BPF_STX_MEM(BPF_W, BPF_REG_10, BPF_REG_0, -4), /* *(u32 *)(fp - 4) = r0 */
+ BPF_MOV64_REG(BPF_REG_2, BPF_REG_10),
+ BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -4), /* r2 = fp - 4 */
+ BPF_LD_MAP_FD(BPF_REG_1, map_fd),
+ BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_map_lookup_elem),
+ BPF_JMP_IMM(BPF_JEQ, BPF_REG_0, 0, 2),
+ BPF_MOV64_IMM(BPF_REG_1, val), /* r1 = 1 */
+ BPF_RAW_INSN(BPF_STX | BPF_XADD | BPF_DW, BPF_REG_0, BPF_REG_1, 0, 0), /* xadd r0 += r1 */
+ BPF_MOV64_IMM(BPF_REG_0, verdict), /* r0 = verdict */
+ BPF_EXIT_INSN(),
+ };
+ size_t insns_cnt = sizeof(prog) / sizeof(struct bpf_insn);
+ int ret;
+
+ ret = bpf_load_program(BPF_PROG_TYPE_CGROUP_SKB,
+ prog, insns_cnt, "GPL", 0,
+ bpf_log_buf, BPF_LOG_BUF_SIZE);
+
+ if (ret < 0) {
+ log_err("Loading program");
+ printf("Output from verifier:\n%s\n-------\n", bpf_log_buf);
+ return 0;
+ }
+ return ret;
+}
+
+
+static int test_multiprog(void)
+{
+ int cg1 = 0, cg2 = 0, cg3 = 0, cg4 = 0, cg5 = 0, key = 0;
+ int drop_prog, allow_prog[6] = {}, rc = 0;
+ unsigned long long value;
+ int i = 0;
+
+ for (i = 0; i < 6; i++) {
+ allow_prog[i] = prog_load_cnt(1, 1 << i);
+ if (!allow_prog[i])
+ goto err;
+ }
+ drop_prog = prog_load_cnt(0, 1);
+ if (!drop_prog)
+ goto err;
+
+ if (setup_cgroup_environment())
+ goto err;
+
+ cg1 = create_and_get_cgroup("/cg1");
+ if (!cg1)
+ goto err;
+ cg2 = create_and_get_cgroup("/cg1/cg2");
+ if (!cg2)
+ goto err;
+ cg3 = create_and_get_cgroup("/cg1/cg2/cg3");
+ if (!cg3)
+ goto err;
+ cg4 = create_and_get_cgroup("/cg1/cg2/cg3/cg4");
+ if (!cg4)
+ goto err;
+ cg5 = create_and_get_cgroup("/cg1/cg2/cg3/cg4/cg5");
+ if (!cg5)
+ goto err;
+
+ if (join_cgroup("/cg1/cg2/cg3/cg4/cg5"))
+ goto err;
+
+ if (bpf_prog_attach(allow_prog[0], cg1, BPF_CGROUP_INET_EGRESS, 2)) {
+ log_err("Attaching prog to cg1");
+ goto err;
+ }
+ if (!bpf_prog_attach(allow_prog[0], cg1, BPF_CGROUP_INET_EGRESS, 2)) {
+ log_err("Unexpected success attaching the same prog to cg1");
+ goto err;
+ }
+ if (bpf_prog_attach(allow_prog[1], cg1, BPF_CGROUP_INET_EGRESS, 2)) {
+ log_err("Attaching prog2 to cg1");
+ goto err;
+ }
+ if (bpf_prog_attach(allow_prog[2], cg2, BPF_CGROUP_INET_EGRESS, 1)) {
+ log_err("Attaching prog to cg2");
+ goto err;
+ }
+ if (bpf_prog_attach(allow_prog[3], cg3, BPF_CGROUP_INET_EGRESS, 2)) {
+ log_err("Attaching prog to cg3");
+ goto err;
+ }
+ if (bpf_prog_attach(allow_prog[4], cg4, BPF_CGROUP_INET_EGRESS, 1)) {
+ log_err("Attaching prog to cg4");
+ goto err;
+ }
+ if (bpf_prog_attach(allow_prog[5], cg5, BPF_CGROUP_INET_EGRESS, 0)) {
+ log_err("Attaching prog to cg5");
+ goto err;
+ }
+ assert(system(PING_CMD) == 0);
+ assert(bpf_map_lookup_elem(map_fd, &key, &value) == 0);
+ assert(value == 1 + 2 + 8 + 32);
+
+ /* detach bottom program and ping again */
+ if (bpf_prog_detach2(-1, cg5, BPF_CGROUP_INET_EGRESS)) {
+ log_err("Detaching prog from cg5");
+ goto err;
+ }
+ value = 0;
+ assert(bpf_map_update_elem(map_fd, &key, &value, 0) == 0);
+ assert(system(PING_CMD) == 0);
+ assert(bpf_map_lookup_elem(map_fd, &key, &value) == 0);
+ assert(value == 1 + 2 + 8 + 16);
+
+ /* detach 3rd from bottom program and ping again */
+ errno = 0;
+ if (!bpf_prog_detach2(0, cg3, BPF_CGROUP_INET_EGRESS)) {
+ log_err("Unexpected success on detach from cg3");
+ goto err;
+ }
+ if (bpf_prog_detach2(allow_prog[3], cg3, BPF_CGROUP_INET_EGRESS)) {
+ log_err("Detaching from cg3");
+ goto err;
+ }
+ value = 0;
+ assert(bpf_map_update_elem(map_fd, &key, &value, 0) == 0);
+ assert(system(PING_CMD) == 0);
+ assert(bpf_map_lookup_elem(map_fd, &key, &value) == 0);
+ assert(value == 1 + 2 + 16);
+
+ /* detach 2nd from bottom program and ping again */
+ if (bpf_prog_detach2(-1, cg4, BPF_CGROUP_INET_EGRESS)) {
+ log_err("Detaching prog from cg4");
+ goto err;
+ }
+ value = 0;
+ assert(bpf_map_update_elem(map_fd, &key, &value, 0) == 0);
+ assert(system(PING_CMD) == 0);
+ assert(bpf_map_lookup_elem(map_fd, &key, &value) == 0);
+ assert(value == 1 + 2 + 4);
+ goto out;
+err:
+ rc = 1;
+
+out:
+ for (i = 0; i < 6; i++)
+ if (allow_prog[i] > 0)
+ close(allow_prog[i]);
+ close(cg1);
+ close(cg2);
+ close(cg3);
+ close(cg4);
+ close(cg5);
+ cleanup_cgroup_environment();
+ if (!rc)
+ printf("### multi:PASS\n");
else
- printf("FAIL\n");
+ printf("### multi:FAIL\n");
return rc;
}
+
+int main(int argc, char **argv)
+{
+ int rc = 0;
+
+ rc = test_foo_bar();
+ if (rc)
+ return rc;
+
+ return test_multiprog();
+}
--
2.9.5
^ permalink raw reply related
* [PATCH net-next 3/8] bpf: enforce return code for cgroup-bpf programs
From: Alexei Starovoitov @ 2017-10-02 23:48 UTC (permalink / raw)
To: David S . Miller
Cc: Daniel Borkmann, Tejun Heo, David Ahern, netdev, kernel-team
In-Reply-To: <20171002234857.3707580-1-ast@fb.com>
with addition of tnum logic the verifier got smart enough and
we can enforce return codes at program load time.
For now do so for cgroup-bpf program types.
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
---
kernel/bpf/verifier.c | 40 ++++++++++++++++
tools/testing/selftests/bpf/test_verifier.c | 72 +++++++++++++++++++++++++++++
2 files changed, 112 insertions(+)
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index 4cf9b72c59a0..52b022310f6a 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -3073,6 +3073,43 @@ static int check_ld_abs(struct bpf_verifier_env *env, struct bpf_insn *insn)
return 0;
}
+static int check_return_code(struct bpf_verifier_env *env)
+{
+ struct bpf_reg_state *reg;
+ struct tnum range = tnum_range(0, 1);
+
+ switch (env->prog->type) {
+ case BPF_PROG_TYPE_CGROUP_SKB:
+ case BPF_PROG_TYPE_CGROUP_SOCK:
+ case BPF_PROG_TYPE_SOCK_OPS:
+ break;
+ default:
+ return 0;
+ }
+
+ reg = &env->cur_state.regs[BPF_REG_0];
+ if (reg->type != SCALAR_VALUE) {
+ verbose("At program exit the register R0 is not a known value (%s)\n",
+ reg_type_str[reg->type]);
+ return -EINVAL;
+ }
+
+ if (!tnum_in(range, reg->var_off)) {
+ verbose("At program exit the register R0 ");
+ if (!tnum_is_unknown(reg->var_off)) {
+ char tn_buf[48];
+
+ tnum_strn(tn_buf, sizeof(tn_buf), reg->var_off);
+ verbose("has value %s", tn_buf);
+ } else {
+ verbose("has unknown scalar value");
+ }
+ verbose(" should have been 0 or 1\n");
+ return -EINVAL;
+ }
+ return 0;
+}
+
/* non-recursive DFS pseudo code
* 1 procedure DFS-iterative(G,v):
* 2 label v as discovered
@@ -3863,6 +3900,9 @@ static int do_check(struct bpf_verifier_env *env)
return -EACCES;
}
+ err = check_return_code(env);
+ if (err)
+ return err;
process_bpf_exit:
insn_idx = pop_stack(env, &prev_insn_idx);
if (insn_idx < 0) {
diff --git a/tools/testing/selftests/bpf/test_verifier.c b/tools/testing/selftests/bpf/test_verifier.c
index 290d5056c165..cc91d0159f43 100644
--- a/tools/testing/selftests/bpf/test_verifier.c
+++ b/tools/testing/selftests/bpf/test_verifier.c
@@ -6892,6 +6892,78 @@ static struct bpf_test tests[] = {
.result = ACCEPT,
.prog_type = BPF_PROG_TYPE_XDP,
},
+ {
+ "bpf_exit with invalid return code. test1",
+ .insns = {
+ BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1, 0),
+ BPF_EXIT_INSN(),
+ },
+ .errstr = "R0 has value (0x0; 0xffffffff)",
+ .result = REJECT,
+ .prog_type = BPF_PROG_TYPE_CGROUP_SOCK,
+ },
+ {
+ "bpf_exit with invalid return code. test2",
+ .insns = {
+ BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1, 0),
+ BPF_ALU64_IMM(BPF_AND, BPF_REG_0, 1),
+ BPF_EXIT_INSN(),
+ },
+ .result = ACCEPT,
+ .prog_type = BPF_PROG_TYPE_CGROUP_SOCK,
+ },
+ {
+ "bpf_exit with invalid return code. test3",
+ .insns = {
+ BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1, 0),
+ BPF_ALU64_IMM(BPF_AND, BPF_REG_0, 3),
+ BPF_EXIT_INSN(),
+ },
+ .errstr = "R0 has value (0x0; 0x3)",
+ .result = REJECT,
+ .prog_type = BPF_PROG_TYPE_CGROUP_SOCK,
+ },
+ {
+ "bpf_exit with invalid return code. test4",
+ .insns = {
+ BPF_MOV64_IMM(BPF_REG_0, 1),
+ BPF_EXIT_INSN(),
+ },
+ .result = ACCEPT,
+ .prog_type = BPF_PROG_TYPE_CGROUP_SOCK,
+ },
+ {
+ "bpf_exit with invalid return code. test5",
+ .insns = {
+ BPF_MOV64_IMM(BPF_REG_0, 2),
+ BPF_EXIT_INSN(),
+ },
+ .errstr = "R0 has value (0x2; 0x0)",
+ .result = REJECT,
+ .prog_type = BPF_PROG_TYPE_CGROUP_SOCK,
+ },
+ {
+ "bpf_exit with invalid return code. test6",
+ .insns = {
+ BPF_MOV64_REG(BPF_REG_0, BPF_REG_1),
+ BPF_EXIT_INSN(),
+ },
+ .errstr = "R0 is not a known value (ctx)",
+ .result = REJECT,
+ .prog_type = BPF_PROG_TYPE_CGROUP_SOCK,
+ },
+ {
+ "bpf_exit with invalid return code. test7",
+ .insns = {
+ BPF_LDX_MEM(BPF_W, BPF_REG_0, BPF_REG_1, 0),
+ BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_1, 4),
+ BPF_ALU64_REG(BPF_MUL, BPF_REG_0, BPF_REG_2),
+ BPF_EXIT_INSN(),
+ },
+ .errstr = "R0 has unknown scalar value",
+ .result = REJECT,
+ .prog_type = BPF_PROG_TYPE_CGROUP_SOCK,
+ },
};
static int probe_filter_length(const struct bpf_insn *fp)
--
2.9.5
^ permalink raw reply related
* [PATCH net-next 4/8] libbpf: introduce bpf_prog_detach2()
From: Alexei Starovoitov @ 2017-10-02 23:48 UTC (permalink / raw)
To: David S . Miller
Cc: Daniel Borkmann, Tejun Heo, David Ahern, netdev, kernel-team
In-Reply-To: <20171002234857.3707580-1-ast@fb.com>
introduce bpf_prog_detach2() that takes one more argument prog_fd
vs bpf_prog_detach() that takes only attach_fd and type.
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
---
tools/lib/bpf/bpf.c | 12 ++++++++++++
tools/lib/bpf/bpf.h | 1 +
2 files changed, 13 insertions(+)
diff --git a/tools/lib/bpf/bpf.c b/tools/lib/bpf/bpf.c
index daf624e4c720..d4b6ba8292ee 100644
--- a/tools/lib/bpf/bpf.c
+++ b/tools/lib/bpf/bpf.c
@@ -291,6 +291,18 @@ int bpf_prog_detach(int target_fd, enum bpf_attach_type type)
return sys_bpf(BPF_PROG_DETACH, &attr, sizeof(attr));
}
+int bpf_prog_detach2(int prog_fd, int target_fd, enum bpf_attach_type type)
+{
+ union bpf_attr attr;
+
+ bzero(&attr, sizeof(attr));
+ attr.target_fd = target_fd;
+ attr.attach_bpf_fd = prog_fd;
+ attr.attach_type = type;
+
+ return sys_bpf(BPF_PROG_DETACH, &attr, sizeof(attr));
+}
+
int bpf_prog_test_run(int prog_fd, int repeat, void *data, __u32 size,
void *data_out, __u32 *size_out, __u32 *retval,
__u32 *duration)
diff --git a/tools/lib/bpf/bpf.h b/tools/lib/bpf/bpf.h
index 118d00535a0d..afd64727c9cf 100644
--- a/tools/lib/bpf/bpf.h
+++ b/tools/lib/bpf/bpf.h
@@ -66,6 +66,7 @@ int bpf_obj_get(const char *pathname);
int bpf_prog_attach(int prog_fd, int attachable_fd, enum bpf_attach_type type,
unsigned int flags);
int bpf_prog_detach(int attachable_fd, enum bpf_attach_type type);
+int bpf_prog_detach2(int prog_fd, int attachable_fd, enum bpf_attach_type type);
int bpf_prog_test_run(int prog_fd, int repeat, void *data, __u32 size,
void *data_out, __u32 *size_out, __u32 *retval,
__u32 *duration);
--
2.9.5
^ permalink raw reply related
* [PATCH net-next 1/8] bpf: multi program support for cgroup+bpf
From: Alexei Starovoitov @ 2017-10-02 23:48 UTC (permalink / raw)
To: David S . Miller
Cc: Daniel Borkmann, Tejun Heo, David Ahern, netdev, kernel-team
In-Reply-To: <20171002234857.3707580-1-ast@fb.com>
introduce BPF_F_ALLOW_MULTI flag that can be used to attach multiple
bpf programs to a cgroup.
The difference between three possible flags for BPF_PROG_ATTACH command:
- NONE(default): No further bpf programs allowed in the subtree.
- BPF_F_ALLOW_OVERRIDE: If a sub-cgroup installs some bpf program,
the program in this cgroup yields to sub-cgroup program.
- BPF_F_ALLOW_MULTI: If a sub-cgroup installs some bpf program,
that cgroup program gets run in addition to the program in this cgroup.
NONE and BPF_F_ALLOW_OVERRIDE existed before. This patch doesn't
change their behavior. It only clarifies the semantics in relation
to new flag.
Only one program is allowed to be attached to a cgroup with
NONE or BPF_F_ALLOW_OVERRIDE flag.
Multiple programs are allowed to be attached to a cgroup with
BPF_F_ALLOW_MULTI flag. They are executed in FIFO order
(those that were attached first, run first)
The programs of sub-cgroup are executed first, then programs of
this cgroup and then programs of parent cgroup.
All eligible programs are executed regardless of return code from
earlier programs.
To allow efficient execution of multiple programs attached to a cgroup
and to avoid penalizing cgroups without any programs attached
introduce 'struct bpf_prog_array' which is RCU protected array
of pointers to bpf programs.
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
for cgroup bits
Acked-by: Tejun Heo <tj@kernel.org>
---
include/linux/bpf-cgroup.h | 46 +++--
include/linux/bpf.h | 32 ++++
include/linux/filter.h | 2 +-
include/uapi/linux/bpf.h | 42 +++-
kernel/bpf/cgroup.c | 467 ++++++++++++++++++++++++++++++++-------------
kernel/bpf/core.c | 31 +++
kernel/bpf/syscall.c | 37 ++--
kernel/cgroup/cgroup.c | 28 ++-
8 files changed, 516 insertions(+), 169 deletions(-)
diff --git a/include/linux/bpf-cgroup.h b/include/linux/bpf-cgroup.h
index d41d40ac3efd..102e56fbb6de 100644
--- a/include/linux/bpf-cgroup.h
+++ b/include/linux/bpf-cgroup.h
@@ -14,27 +14,42 @@ struct bpf_sock_ops_kern;
extern struct static_key_false cgroup_bpf_enabled_key;
#define cgroup_bpf_enabled static_branch_unlikely(&cgroup_bpf_enabled_key)
+struct bpf_prog_list {
+ struct list_head node;
+ struct bpf_prog *prog;
+};
+
+struct bpf_prog_array;
+
struct cgroup_bpf {
- /*
- * Store two sets of bpf_prog pointers, one for programs that are
- * pinned directly to this cgroup, and one for those that are effective
- * when this cgroup is accessed.
+ /* array of effective progs in this cgroup */
+ struct bpf_prog_array __rcu *effective[MAX_BPF_ATTACH_TYPE];
+
+ /* attached progs to this cgroup and attach flags
+ * when flags == 0 or BPF_F_ALLOW_OVERRIDE the progs list will
+ * have either zero or one element
+ * when BPF_F_ALLOW_MULTI the list can have up to BPF_CGROUP_MAX_PROGS
*/
- struct bpf_prog *prog[MAX_BPF_ATTACH_TYPE];
- struct bpf_prog __rcu *effective[MAX_BPF_ATTACH_TYPE];
- bool disallow_override[MAX_BPF_ATTACH_TYPE];
+ struct list_head progs[MAX_BPF_ATTACH_TYPE];
+ u32 flags[MAX_BPF_ATTACH_TYPE];
+
+ /* temp storage for effective prog array used by prog_attach/detach */
+ struct bpf_prog_array __rcu *inactive;
};
void cgroup_bpf_put(struct cgroup *cgrp);
-void cgroup_bpf_inherit(struct cgroup *cgrp, struct cgroup *parent);
+int cgroup_bpf_inherit(struct cgroup *cgrp);
-int __cgroup_bpf_update(struct cgroup *cgrp, struct cgroup *parent,
- struct bpf_prog *prog, enum bpf_attach_type type,
- bool overridable);
+int __cgroup_bpf_attach(struct cgroup *cgrp, struct bpf_prog *prog,
+ enum bpf_attach_type type, u32 flags);
+int __cgroup_bpf_detach(struct cgroup *cgrp, struct bpf_prog *prog,
+ enum bpf_attach_type type, u32 flags);
-/* Wrapper for __cgroup_bpf_update() protected by cgroup_mutex */
-int cgroup_bpf_update(struct cgroup *cgrp, struct bpf_prog *prog,
- enum bpf_attach_type type, bool overridable);
+/* Wrapper for __cgroup_bpf_*() protected by cgroup_mutex */
+int cgroup_bpf_attach(struct cgroup *cgrp, struct bpf_prog *prog,
+ enum bpf_attach_type type, u32 flags);
+int cgroup_bpf_detach(struct cgroup *cgrp, struct bpf_prog *prog,
+ enum bpf_attach_type type, u32 flags);
int __cgroup_bpf_run_filter_skb(struct sock *sk,
struct sk_buff *skb,
@@ -96,8 +111,7 @@ int __cgroup_bpf_run_filter_sock_ops(struct sock *sk,
struct cgroup_bpf {};
static inline void cgroup_bpf_put(struct cgroup *cgrp) {}
-static inline void cgroup_bpf_inherit(struct cgroup *cgrp,
- struct cgroup *parent) {}
+static inline int cgroup_bpf_inherit(struct cgroup *cgrp) { return 0; }
#define BPF_CGROUP_RUN_PROG_INET_INGRESS(sk,skb) ({ 0; })
#define BPF_CGROUP_RUN_PROG_INET_EGRESS(sk,skb) ({ 0; })
diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 252f4bc9eb25..a6964b75f070 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -241,6 +241,38 @@ int bpf_prog_test_run_xdp(struct bpf_prog *prog, const union bpf_attr *kattr,
int bpf_prog_test_run_skb(struct bpf_prog *prog, const union bpf_attr *kattr,
union bpf_attr __user *uattr);
+/* an array of programs to be executed under rcu_lock.
+ *
+ * Typical usage:
+ * ret = BPF_PROG_RUN_ARRAY(&bpf_prog_array, ctx, BPF_PROG_RUN);
+ *
+ * the structure returned by bpf_prog_array_alloc() should be populated
+ * with program pointers and the last pointer must be NULL.
+ * The user has to keep refcnt on the program and make sure the program
+ * is removed from the array before bpf_prog_put().
+ * The 'struct bpf_prog_array *' should only be replaced with xchg()
+ * since other cpus are walking the array of pointers in parallel.
+ */
+struct bpf_prog_array {
+ struct rcu_head rcu;
+ struct bpf_prog *progs[0];
+};
+
+struct bpf_prog_array __rcu *bpf_prog_array_alloc(u32 prog_cnt, gfp_t flags);
+void bpf_prog_array_free(struct bpf_prog_array __rcu *progs);
+
+#define BPF_PROG_RUN_ARRAY(array, ctx, func) \
+ ({ \
+ struct bpf_prog **_prog; \
+ u32 _ret = 1; \
+ rcu_read_lock(); \
+ _prog = rcu_dereference(array)->progs; \
+ for (; *_prog; _prog++) \
+ _ret &= func(*_prog, ctx); \
+ rcu_read_unlock(); \
+ _ret; \
+ })
+
#ifdef CONFIG_BPF_SYSCALL
DECLARE_PER_CPU(int, bpf_prog_active);
diff --git a/include/linux/filter.h b/include/linux/filter.h
index 911d454af107..2d2db394b0ca 100644
--- a/include/linux/filter.h
+++ b/include/linux/filter.h
@@ -481,7 +481,7 @@ struct sk_filter {
struct bpf_prog *prog;
};
-#define BPF_PROG_RUN(filter, ctx) (*filter->bpf_func)(ctx, filter->insnsi)
+#define BPF_PROG_RUN(filter, ctx) (*(filter)->bpf_func)(ctx, (filter)->insnsi)
#define BPF_SKB_CB_LEN QDISC_CB_PRIV_LEN
diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index 6d2137b4cf38..762f74bc6c47 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -143,11 +143,47 @@ enum bpf_attach_type {
#define MAX_BPF_ATTACH_TYPE __MAX_BPF_ATTACH_TYPE
-/* If BPF_F_ALLOW_OVERRIDE flag is used in BPF_PROG_ATTACH command
- * to the given target_fd cgroup the descendent cgroup will be able to
- * override effective bpf program that was inherited from this cgroup
+/* cgroup-bpf attach flags used in BPF_PROG_ATTACH command
+ *
+ * NONE(default): No further bpf programs allowed in the subtree.
+ *
+ * BPF_F_ALLOW_OVERRIDE: If a sub-cgroup installs some bpf program,
+ * the program in this cgroup yields to sub-cgroup program.
+ *
+ * BPF_F_ALLOW_MULTI: If a sub-cgroup installs some bpf program,
+ * that cgroup program gets run in addition to the program in this cgroup.
+ *
+ * Only one program is allowed to be attached to a cgroup with
+ * NONE or BPF_F_ALLOW_OVERRIDE flag.
+ * Attaching another program on top of NONE or BPF_F_ALLOW_OVERRIDE will
+ * release old program and attach the new one. Attach flags has to match.
+ *
+ * Multiple programs are allowed to be attached to a cgroup with
+ * BPF_F_ALLOW_MULTI flag. They are executed in FIFO order
+ * (those that were attached first, run first)
+ * The programs of sub-cgroup are executed first, then programs of
+ * this cgroup and then programs of parent cgroup.
+ * When children program makes decision (like picking TCP CA or sock bind)
+ * parent program has a chance to override it.
+ *
+ * A cgroup with MULTI or OVERRIDE flag allows any attach flags in sub-cgroups.
+ * A cgroup with NONE doesn't allow any programs in sub-cgroups.
+ * Ex1:
+ * cgrp1 (MULTI progs A, B) ->
+ * cgrp2 (OVERRIDE prog C) ->
+ * cgrp3 (MULTI prog D) ->
+ * cgrp4 (OVERRIDE prog E) ->
+ * cgrp5 (NONE prog F)
+ * the event in cgrp5 triggers execution of F,D,A,B in that order.
+ * if prog F is detached, the execution is E,D,A,B
+ * if prog F and D are detached, the execution is E,A,B
+ * if prog F, E and D are detached, the execution is C,A,B
+ *
+ * All eligible programs are executed regardless of return code from
+ * earlier programs.
*/
#define BPF_F_ALLOW_OVERRIDE (1U << 0)
+#define BPF_F_ALLOW_MULTI (1U << 1)
/* If BPF_F_STRICT_ALIGNMENT is used in BPF_PROG_LOAD command, the
* verifier will perform strict alignment checking as if the kernel
diff --git a/kernel/bpf/cgroup.c b/kernel/bpf/cgroup.c
index 546113430049..70f679a94804 100644
--- a/kernel/bpf/cgroup.c
+++ b/kernel/bpf/cgroup.c
@@ -27,129 +27,361 @@ void cgroup_bpf_put(struct cgroup *cgrp)
{
unsigned int type;
- for (type = 0; type < ARRAY_SIZE(cgrp->bpf.prog); type++) {
- struct bpf_prog *prog = cgrp->bpf.prog[type];
-
- if (prog) {
- bpf_prog_put(prog);
+ for (type = 0; type < ARRAY_SIZE(cgrp->bpf.progs); type++) {
+ struct list_head *progs = &cgrp->bpf.progs[type];
+ struct bpf_prog_list *pl, *tmp;
+
+ list_for_each_entry_safe(pl, tmp, progs, node) {
+ list_del(&pl->node);
+ bpf_prog_put(pl->prog);
+ kfree(pl);
static_branch_dec(&cgroup_bpf_enabled_key);
}
+ bpf_prog_array_free(cgrp->bpf.effective[type]);
+ }
+}
+
+/* count number of elements in the list.
+ * it's slow but the list cannot be long
+ */
+static u32 prog_list_length(struct list_head *head)
+{
+ struct bpf_prog_list *pl;
+ u32 cnt = 0;
+
+ list_for_each_entry(pl, head, node) {
+ if (!pl->prog)
+ continue;
+ cnt++;
}
+ return cnt;
+}
+
+/* if parent has non-overridable prog attached,
+ * disallow attaching new programs to the descendent cgroup.
+ * if parent has overridable or multi-prog, allow attaching
+ */
+static bool hierarchy_allows_attach(struct cgroup *cgrp,
+ enum bpf_attach_type type,
+ u32 new_flags)
+{
+ struct cgroup *p;
+
+ p = cgroup_parent(cgrp);
+ if (!p)
+ return true;
+ do {
+ u32 flags = p->bpf.flags[type];
+ u32 cnt;
+
+ if (flags & BPF_F_ALLOW_MULTI)
+ return true;
+ cnt = prog_list_length(&p->bpf.progs[type]);
+ WARN_ON_ONCE(cnt > 1);
+ if (cnt == 1)
+ return !!(flags & BPF_F_ALLOW_OVERRIDE);
+ p = cgroup_parent(p);
+ } while (p);
+ return true;
+}
+
+/* compute a chain of effective programs for a given cgroup:
+ * start from the list of programs in this cgroup and add
+ * all parent programs.
+ * Note that parent's F_ALLOW_OVERRIDE-type program is yielding
+ * to programs in this cgroup
+ */
+static int compute_effective_progs(struct cgroup *cgrp,
+ enum bpf_attach_type type,
+ struct bpf_prog_array __rcu **array)
+{
+ struct bpf_prog_array __rcu *progs;
+ struct bpf_prog_list *pl;
+ struct cgroup *p = cgrp;
+ int cnt = 0;
+
+ /* count number of effective programs by walking parents */
+ do {
+ if (cnt == 0 || (p->bpf.flags[type] & BPF_F_ALLOW_MULTI))
+ cnt += prog_list_length(&p->bpf.progs[type]);
+ p = cgroup_parent(p);
+ } while (p);
+
+ progs = bpf_prog_array_alloc(cnt, GFP_KERNEL);
+ if (!progs)
+ return -ENOMEM;
+
+ /* populate the array with effective progs */
+ cnt = 0;
+ p = cgrp;
+ do {
+ if (cnt == 0 || (p->bpf.flags[type] & BPF_F_ALLOW_MULTI))
+ list_for_each_entry(pl,
+ &p->bpf.progs[type], node) {
+ if (!pl->prog)
+ continue;
+ rcu_dereference_protected(progs, 1)->
+ progs[cnt++] = pl->prog;
+ }
+ p = cgroup_parent(p);
+ } while (p);
+
+ *array = progs;
+ return 0;
+}
+
+static void activate_effective_progs(struct cgroup *cgrp,
+ enum bpf_attach_type type,
+ struct bpf_prog_array __rcu *array)
+{
+ struct bpf_prog_array __rcu *old_array;
+
+ old_array = xchg(&cgrp->bpf.effective[type], array);
+ /* free prog array after grace period, since __cgroup_bpf_run_*()
+ * might be still walking the array
+ */
+ bpf_prog_array_free(old_array);
}
/**
* cgroup_bpf_inherit() - inherit effective programs from parent
* @cgrp: the cgroup to modify
- * @parent: the parent to inherit from
*/
-void cgroup_bpf_inherit(struct cgroup *cgrp, struct cgroup *parent)
+int cgroup_bpf_inherit(struct cgroup *cgrp)
{
- unsigned int type;
+/* has to use marco instead of const int, since compiler thinks
+ * that array below is variable length
+ */
+#define NR ARRAY_SIZE(cgrp->bpf.effective)
+ struct bpf_prog_array __rcu *arrays[NR] = {};
+ int i;
- for (type = 0; type < ARRAY_SIZE(cgrp->bpf.effective); type++) {
- struct bpf_prog *e;
+ for (i = 0; i < NR; i++)
+ INIT_LIST_HEAD(&cgrp->bpf.progs[i]);
- e = rcu_dereference_protected(parent->bpf.effective[type],
- lockdep_is_held(&cgroup_mutex));
- rcu_assign_pointer(cgrp->bpf.effective[type], e);
- cgrp->bpf.disallow_override[type] = parent->bpf.disallow_override[type];
- }
+ for (i = 0; i < NR; i++)
+ if (compute_effective_progs(cgrp, i, &arrays[i]))
+ goto cleanup;
+
+ for (i = 0; i < NR; i++)
+ activate_effective_progs(cgrp, i, arrays[i]);
+
+ return 0;
+cleanup:
+ for (i = 0; i < NR; i++)
+ bpf_prog_array_free(arrays[i]);
+ return -ENOMEM;
}
+#define BPF_CGROUP_MAX_PROGS 64
+
/**
- * __cgroup_bpf_update() - Update the pinned program of a cgroup, and
+ * __cgroup_bpf_attach() - Attach the program to a cgroup, and
* propagate the change to descendants
* @cgrp: The cgroup which descendants to traverse
- * @parent: The parent of @cgrp, or %NULL if @cgrp is the root
- * @prog: A new program to pin
- * @type: Type of pinning operation (ingress/egress)
- *
- * Each cgroup has a set of two pointers for bpf programs; one for eBPF
- * programs it owns, and which is effective for execution.
- *
- * If @prog is not %NULL, this function attaches a new program to the cgroup
- * and releases the one that is currently attached, if any. @prog is then made
- * the effective program of type @type in that cgroup.
- *
- * If @prog is %NULL, the currently attached program of type @type is released,
- * and the effective program of the parent cgroup (if any) is inherited to
- * @cgrp.
- *
- * Then, the descendants of @cgrp are walked and the effective program for
- * each of them is set to the effective program of @cgrp unless the
- * descendant has its own program attached, in which case the subbranch is
- * skipped. This ensures that delegated subcgroups with own programs are left
- * untouched.
+ * @prog: A program to attach
+ * @type: Type of attach operation
*
* Must be called with cgroup_mutex held.
*/
-int __cgroup_bpf_update(struct cgroup *cgrp, struct cgroup *parent,
- struct bpf_prog *prog, enum bpf_attach_type type,
- bool new_overridable)
+int __cgroup_bpf_attach(struct cgroup *cgrp, struct bpf_prog *prog,
+ enum bpf_attach_type type, u32 flags)
{
- struct bpf_prog *old_prog, *effective = NULL;
- struct cgroup_subsys_state *pos;
- bool overridable = true;
-
- if (parent) {
- overridable = !parent->bpf.disallow_override[type];
- effective = rcu_dereference_protected(parent->bpf.effective[type],
- lockdep_is_held(&cgroup_mutex));
- }
-
- if (prog && effective && !overridable)
- /* if parent has non-overridable prog attached, disallow
- * attaching new programs to descendent cgroup
- */
+ struct list_head *progs = &cgrp->bpf.progs[type];
+ struct bpf_prog *old_prog = NULL;
+ struct cgroup_subsys_state *css;
+ struct bpf_prog_list *pl;
+ bool pl_was_allocated;
+ u32 old_flags;
+ int err;
+
+ if ((flags & BPF_F_ALLOW_OVERRIDE) && (flags & BPF_F_ALLOW_MULTI))
+ /* invalid combination */
+ return -EINVAL;
+
+ if (!hierarchy_allows_attach(cgrp, type, flags))
return -EPERM;
- if (prog && effective && overridable != new_overridable)
- /* if parent has overridable prog attached, only
- * allow overridable programs in descendent cgroup
+ if (!list_empty(progs) && cgrp->bpf.flags[type] != flags)
+ /* Disallow attaching non-overridable on top
+ * of existing overridable in this cgroup.
+ * Disallow attaching multi-prog if overridable or none
*/
return -EPERM;
- old_prog = cgrp->bpf.prog[type];
-
- if (prog) {
- overridable = new_overridable;
- effective = prog;
- if (old_prog &&
- cgrp->bpf.disallow_override[type] == new_overridable)
- /* disallow attaching non-overridable on top
- * of existing overridable in this cgroup
- * and vice versa
- */
- return -EPERM;
+ if (prog_list_length(progs) >= BPF_CGROUP_MAX_PROGS)
+ return -E2BIG;
+
+ if (flags & BPF_F_ALLOW_MULTI) {
+ list_for_each_entry(pl, progs, node)
+ if (pl->prog == prog)
+ /* disallow attaching the same prog twice */
+ return -EINVAL;
+
+ pl = kmalloc(sizeof(*pl), GFP_KERNEL);
+ if (!pl)
+ return -ENOMEM;
+ pl_was_allocated = true;
+ pl->prog = prog;
+ list_add_tail(&pl->node, progs);
+ } else {
+ if (list_empty(progs)) {
+ pl = kmalloc(sizeof(*pl), GFP_KERNEL);
+ if (!pl)
+ return -ENOMEM;
+ pl_was_allocated = true;
+ list_add_tail(&pl->node, progs);
+ } else {
+ old_prog = pl->prog;
+ pl = list_first_entry(progs, typeof(*pl), node);
+ pl_was_allocated = false;
+ }
+ pl->prog = prog;
}
- if (!prog && !old_prog)
- /* report error when trying to detach and nothing is attached */
- return -ENOENT;
+ old_flags = cgrp->bpf.flags[type];
+ cgrp->bpf.flags[type] = flags;
- cgrp->bpf.prog[type] = prog;
+ /* allocate and recompute effective prog arrays */
+ css_for_each_descendant_pre(css, &cgrp->self) {
+ struct cgroup *desc = container_of(css, struct cgroup, self);
- css_for_each_descendant_pre(pos, &cgrp->self) {
- struct cgroup *desc = container_of(pos, struct cgroup, self);
-
- /* skip the subtree if the descendant has its own program */
- if (desc->bpf.prog[type] && desc != cgrp) {
- pos = css_rightmost_descendant(pos);
- } else {
- rcu_assign_pointer(desc->bpf.effective[type],
- effective);
- desc->bpf.disallow_override[type] = !overridable;
- }
+ err = compute_effective_progs(desc, type, &desc->bpf.inactive);
+ if (err)
+ goto cleanup;
}
- if (prog)
- static_branch_inc(&cgroup_bpf_enabled_key);
+ /* all allocations were successful. Activate all prog arrays */
+ css_for_each_descendant_pre(css, &cgrp->self) {
+ struct cgroup *desc = container_of(css, struct cgroup, self);
+ activate_effective_progs(desc, type, desc->bpf.inactive);
+ desc->bpf.inactive = NULL;
+ }
+
+ static_branch_inc(&cgroup_bpf_enabled_key);
if (old_prog) {
bpf_prog_put(old_prog);
static_branch_dec(&cgroup_bpf_enabled_key);
}
return 0;
+
+cleanup:
+ /* oom while computing effective. Free all computed effective arrays
+ * since they were not activated
+ */
+ css_for_each_descendant_pre(css, &cgrp->self) {
+ struct cgroup *desc = container_of(css, struct cgroup, self);
+
+ bpf_prog_array_free(desc->bpf.inactive);
+ desc->bpf.inactive = NULL;
+ }
+
+ /* and cleanup the prog list */
+ pl->prog = old_prog;
+ if (pl_was_allocated) {
+ list_del(&pl->node);
+ kfree(pl);
+ }
+ return err;
+}
+
+/**
+ * __cgroup_bpf_detach() - Detach the program from a cgroup, and
+ * propagate the change to descendants
+ * @cgrp: The cgroup which descendants to traverse
+ * @prog: A program to detach or NULL
+ * @type: Type of detach operation
+ *
+ * Must be called with cgroup_mutex held.
+ */
+int __cgroup_bpf_detach(struct cgroup *cgrp, struct bpf_prog *prog,
+ enum bpf_attach_type type, u32 unused_flags)
+{
+ struct list_head *progs = &cgrp->bpf.progs[type];
+ u32 flags = cgrp->bpf.flags[type];
+ struct bpf_prog *old_prog = NULL;
+ struct cgroup_subsys_state *css;
+ struct bpf_prog_list *pl;
+ int err;
+
+ if (flags & BPF_F_ALLOW_MULTI) {
+ if (!prog)
+ /* to detach MULTI prog the user has to specify valid FD
+ * of the program to be detached
+ */
+ return -EINVAL;
+ } else {
+ if (list_empty(progs))
+ /* report error when trying to detach and nothing is attached */
+ return -ENOENT;
+ }
+
+ if (flags & BPF_F_ALLOW_MULTI) {
+ /* find the prog and detach it */
+ list_for_each_entry(pl, progs, node) {
+ if (pl->prog != prog)
+ continue;
+ old_prog = prog;
+ /* mark it deleted, so it's ignored while
+ * recomputing effective
+ */
+ pl->prog = NULL;
+ break;
+ }
+ if (!old_prog)
+ return -ENOENT;
+ } else {
+ /* to maintain backward compatibility NONE and OVERRIDE cgroups
+ * allow detaching with invalid FD (prog==NULL)
+ */
+ pl = list_first_entry(progs, typeof(*pl), node);
+ old_prog = pl->prog;
+ pl->prog = NULL;
+ }
+
+ /* allocate and recompute effective prog arrays */
+ css_for_each_descendant_pre(css, &cgrp->self) {
+ struct cgroup *desc = container_of(css, struct cgroup, self);
+
+ err = compute_effective_progs(desc, type, &desc->bpf.inactive);
+ if (err)
+ goto cleanup;
+ }
+
+ /* all allocations were successful. Activate all prog arrays */
+ css_for_each_descendant_pre(css, &cgrp->self) {
+ struct cgroup *desc = container_of(css, struct cgroup, self);
+
+ activate_effective_progs(desc, type, desc->bpf.inactive);
+ desc->bpf.inactive = NULL;
+ }
+
+ /* now can actually delete it from this cgroup list */
+ list_del(&pl->node);
+ kfree(pl);
+ if (list_empty(progs))
+ /* last program was detached, reset flags to zero */
+ cgrp->bpf.flags[type] = 0;
+
+ bpf_prog_put(old_prog);
+ static_branch_dec(&cgroup_bpf_enabled_key);
+ return 0;
+
+cleanup:
+ /* oom while computing effective. Free all computed effective arrays
+ * since they were not activated
+ */
+ css_for_each_descendant_pre(css, &cgrp->self) {
+ struct cgroup *desc = container_of(css, struct cgroup, self);
+
+ bpf_prog_array_free(desc->bpf.inactive);
+ desc->bpf.inactive = NULL;
+ }
+
+ /* and restore back old_prog */
+ pl->prog = old_prog;
+ return err;
}
/**
@@ -171,36 +403,26 @@ int __cgroup_bpf_run_filter_skb(struct sock *sk,
struct sk_buff *skb,
enum bpf_attach_type type)
{
- struct bpf_prog *prog;
+ unsigned int offset = skb->data - skb_network_header(skb);
+ struct sock *save_sk;
struct cgroup *cgrp;
- int ret = 0;
+ int ret;
if (!sk || !sk_fullsock(sk))
return 0;
- if (sk->sk_family != AF_INET &&
- sk->sk_family != AF_INET6)
+ if (sk->sk_family != AF_INET && sk->sk_family != AF_INET6)
return 0;
cgrp = sock_cgroup_ptr(&sk->sk_cgrp_data);
-
- rcu_read_lock();
-
- prog = rcu_dereference(cgrp->bpf.effective[type]);
- if (prog) {
- unsigned int offset = skb->data - skb_network_header(skb);
- struct sock *save_sk = skb->sk;
-
- skb->sk = sk;
- __skb_push(skb, offset);
- ret = bpf_prog_run_save_cb(prog, skb) == 1 ? 0 : -EPERM;
- __skb_pull(skb, offset);
- skb->sk = save_sk;
- }
-
- rcu_read_unlock();
-
- return ret;
+ save_sk = skb->sk;
+ skb->sk = sk;
+ __skb_push(skb, offset);
+ ret = BPF_PROG_RUN_ARRAY(cgrp->bpf.effective[type], skb,
+ bpf_prog_run_save_cb);
+ __skb_pull(skb, offset);
+ skb->sk = save_sk;
+ return ret == 1 ? 0 : -EPERM;
}
EXPORT_SYMBOL(__cgroup_bpf_run_filter_skb);
@@ -221,19 +443,10 @@ int __cgroup_bpf_run_filter_sk(struct sock *sk,
enum bpf_attach_type type)
{
struct cgroup *cgrp = sock_cgroup_ptr(&sk->sk_cgrp_data);
- struct bpf_prog *prog;
- int ret = 0;
-
-
- rcu_read_lock();
-
- prog = rcu_dereference(cgrp->bpf.effective[type]);
- if (prog)
- ret = BPF_PROG_RUN(prog, sk) == 1 ? 0 : -EPERM;
+ int ret;
- rcu_read_unlock();
-
- return ret;
+ ret = BPF_PROG_RUN_ARRAY(cgrp->bpf.effective[type], sk, BPF_PROG_RUN);
+ return ret == 1 ? 0 : -EPERM;
}
EXPORT_SYMBOL(__cgroup_bpf_run_filter_sk);
@@ -258,18 +471,10 @@ int __cgroup_bpf_run_filter_sock_ops(struct sock *sk,
enum bpf_attach_type type)
{
struct cgroup *cgrp = sock_cgroup_ptr(&sk->sk_cgrp_data);
- struct bpf_prog *prog;
- int ret = 0;
-
-
- rcu_read_lock();
-
- prog = rcu_dereference(cgrp->bpf.effective[type]);
- if (prog)
- ret = BPF_PROG_RUN(prog, sock_ops) == 1 ? 0 : -EPERM;
-
- rcu_read_unlock();
+ int ret;
- return ret;
+ ret = BPF_PROG_RUN_ARRAY(cgrp->bpf.effective[type], sock_ops,
+ BPF_PROG_RUN);
+ return ret == 1 ? 0 : -EPERM;
}
EXPORT_SYMBOL(__cgroup_bpf_run_filter_sock_ops);
diff --git a/kernel/bpf/core.c b/kernel/bpf/core.c
index 917cc04a0a94..6b49e1991ae7 100644
--- a/kernel/bpf/core.c
+++ b/kernel/bpf/core.c
@@ -1381,6 +1381,37 @@ struct bpf_prog *bpf_prog_select_runtime(struct bpf_prog *fp, int *err)
}
EXPORT_SYMBOL_GPL(bpf_prog_select_runtime);
+/* to avoid allocating empty bpf_prog_array for cgroups that
+ * don't have bpf program attached use one global 'empty_prog_array'
+ * It will not be modified the caller of bpf_prog_array_alloc()
+ * (since caller requested prog_cnt == 0)
+ * that pointer should be 'freed' by bpf_prog_array_free()
+ */
+static struct {
+ struct bpf_prog_array hdr;
+ struct bpf_prog *null_prog;
+} empty_prog_array = {
+ .null_prog = NULL,
+};
+
+struct bpf_prog_array __rcu *bpf_prog_array_alloc(u32 prog_cnt, gfp_t flags)
+{
+ if (prog_cnt)
+ return kzalloc(sizeof(struct bpf_prog_array) +
+ sizeof(struct bpf_prog *) * (prog_cnt + 1),
+ flags);
+
+ return &empty_prog_array.hdr;
+}
+
+void bpf_prog_array_free(struct bpf_prog_array __rcu *progs)
+{
+ if (!progs ||
+ progs == (struct bpf_prog_array __rcu *)&empty_prog_array.hdr)
+ return;
+ kfree_rcu(progs, rcu);
+}
+
static void bpf_prog_free_deferred(struct work_struct *work)
{
struct bpf_prog_aux *aux;
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index b927da66f653..51bee695d32c 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -1168,6 +1168,9 @@ static int sockmap_get_from_fd(const union bpf_attr *attr, bool attach)
return 0;
}
+#define BPF_F_ATTACH_MASK \
+ (BPF_F_ALLOW_OVERRIDE | BPF_F_ALLOW_MULTI)
+
static int bpf_prog_attach(const union bpf_attr *attr)
{
enum bpf_prog_type ptype;
@@ -1181,7 +1184,7 @@ static int bpf_prog_attach(const union bpf_attr *attr)
if (CHECK_ATTR(BPF_PROG_ATTACH))
return -EINVAL;
- if (attr->attach_flags & ~BPF_F_ALLOW_OVERRIDE)
+ if (attr->attach_flags & ~BPF_F_ATTACH_MASK)
return -EINVAL;
switch (attr->attach_type) {
@@ -1212,8 +1215,8 @@ static int bpf_prog_attach(const union bpf_attr *attr)
return PTR_ERR(cgrp);
}
- ret = cgroup_bpf_update(cgrp, prog, attr->attach_type,
- attr->attach_flags & BPF_F_ALLOW_OVERRIDE);
+ ret = cgroup_bpf_attach(cgrp, prog, attr->attach_type,
+ attr->attach_flags);
if (ret)
bpf_prog_put(prog);
cgroup_put(cgrp);
@@ -1225,6 +1228,8 @@ static int bpf_prog_attach(const union bpf_attr *attr)
static int bpf_prog_detach(const union bpf_attr *attr)
{
+ enum bpf_prog_type ptype;
+ struct bpf_prog *prog;
struct cgroup *cgrp;
int ret;
@@ -1237,23 +1242,33 @@ static int bpf_prog_detach(const union bpf_attr *attr)
switch (attr->attach_type) {
case BPF_CGROUP_INET_INGRESS:
case BPF_CGROUP_INET_EGRESS:
+ ptype = BPF_PROG_TYPE_CGROUP_SKB;
+ break;
case BPF_CGROUP_INET_SOCK_CREATE:
+ ptype = BPF_PROG_TYPE_CGROUP_SOCK;
+ break;
case BPF_CGROUP_SOCK_OPS:
- cgrp = cgroup_get_from_fd(attr->target_fd);
- if (IS_ERR(cgrp))
- return PTR_ERR(cgrp);
-
- ret = cgroup_bpf_update(cgrp, NULL, attr->attach_type, false);
- cgroup_put(cgrp);
+ ptype = BPF_PROG_TYPE_SOCK_OPS;
break;
case BPF_SK_SKB_STREAM_PARSER:
case BPF_SK_SKB_STREAM_VERDICT:
- ret = sockmap_get_from_fd(attr, false);
- break;
+ return sockmap_get_from_fd(attr, false);
default:
return -EINVAL;
}
+ cgrp = cgroup_get_from_fd(attr->target_fd);
+ if (IS_ERR(cgrp))
+ return PTR_ERR(cgrp);
+
+ prog = bpf_prog_get_type(attr->attach_bpf_fd, ptype);
+ if (IS_ERR(prog))
+ prog = NULL;
+
+ ret = cgroup_bpf_detach(cgrp, prog, attr->attach_type, 0);
+ if (prog)
+ bpf_prog_put(prog);
+ cgroup_put(cgrp);
return ret;
}
diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c
index d6551cd45238..57eb866ae78d 100644
--- a/kernel/cgroup/cgroup.c
+++ b/kernel/cgroup/cgroup.c
@@ -1896,6 +1896,9 @@ int cgroup_setup_root(struct cgroup_root *root, u16 ss_mask, int ref_flags)
if (ret)
goto destroy_root;
+ ret = cgroup_bpf_inherit(root_cgrp);
+ WARN_ON_ONCE(ret);
+
trace_cgroup_setup_root(root);
/*
@@ -4713,6 +4716,9 @@ static struct cgroup *cgroup_create(struct cgroup *parent)
cgrp->self.parent = &parent->self;
cgrp->root = root;
cgrp->level = level;
+ ret = cgroup_bpf_inherit(cgrp);
+ if (ret)
+ goto out_idr_free;
for (tcgrp = cgrp; tcgrp; tcgrp = cgroup_parent(tcgrp)) {
cgrp->ancestor_ids[tcgrp->level] = tcgrp->id;
@@ -4747,13 +4753,12 @@ static struct cgroup *cgroup_create(struct cgroup *parent)
if (!cgroup_on_dfl(cgrp))
cgrp->subtree_control = cgroup_control(cgrp);
- if (parent)
- cgroup_bpf_inherit(cgrp, parent);
-
cgroup_propagate_control(cgrp);
return cgrp;
+out_idr_free:
+ cgroup_idr_remove(&root->cgroup_idr, cgrp->id);
out_cancel_ref:
percpu_ref_exit(&cgrp->self.refcnt);
out_free_cgrp:
@@ -5736,14 +5741,23 @@ void cgroup_sk_free(struct sock_cgroup_data *skcd)
#endif /* CONFIG_SOCK_CGROUP_DATA */
#ifdef CONFIG_CGROUP_BPF
-int cgroup_bpf_update(struct cgroup *cgrp, struct bpf_prog *prog,
- enum bpf_attach_type type, bool overridable)
+int cgroup_bpf_attach(struct cgroup *cgrp, struct bpf_prog *prog,
+ enum bpf_attach_type type, u32 flags)
+{
+ int ret;
+
+ mutex_lock(&cgroup_mutex);
+ ret = __cgroup_bpf_attach(cgrp, prog, type, flags);
+ mutex_unlock(&cgroup_mutex);
+ return ret;
+}
+int cgroup_bpf_detach(struct cgroup *cgrp, struct bpf_prog *prog,
+ enum bpf_attach_type type, u32 flags)
{
- struct cgroup *parent = cgroup_parent(cgrp);
int ret;
mutex_lock(&cgroup_mutex);
- ret = __cgroup_bpf_update(cgrp, parent, prog, type, overridable);
+ ret = __cgroup_bpf_detach(cgrp, prog, type, flags);
mutex_unlock(&cgroup_mutex);
return ret;
}
--
2.9.5
^ permalink raw reply related
* [PATCH net-next 2/8] bpf: introduce BPF_PROG_QUERY command
From: Alexei Starovoitov @ 2017-10-02 23:48 UTC (permalink / raw)
To: David S . Miller
Cc: Daniel Borkmann, Tejun Heo, David Ahern, netdev, kernel-team
In-Reply-To: <20171002234857.3707580-1-ast@fb.com>
introduce BPF_PROG_QUERY command to retrieve a set of either
attached programs to given cgroup or a set of effective programs
that will execute for events within a cgroup
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Daniel Borkmann <daniel@iogearbox.net>
for cgroup bits
Acked-by: Tejun Heo <tj@kernel.org>
---
include/linux/bpf-cgroup.h | 4 ++++
include/linux/bpf.h | 3 +++
include/uapi/linux/bpf.h | 13 +++++++++++++
kernel/bpf/cgroup.c | 43 +++++++++++++++++++++++++++++++++++++++++++
kernel/bpf/core.c | 33 +++++++++++++++++++++++++++++++++
kernel/bpf/syscall.c | 34 ++++++++++++++++++++++++++++++++++
kernel/cgroup/cgroup.c | 10 ++++++++++
7 files changed, 140 insertions(+)
diff --git a/include/linux/bpf-cgroup.h b/include/linux/bpf-cgroup.h
index 102e56fbb6de..359b6f5d3d90 100644
--- a/include/linux/bpf-cgroup.h
+++ b/include/linux/bpf-cgroup.h
@@ -44,12 +44,16 @@ int __cgroup_bpf_attach(struct cgroup *cgrp, struct bpf_prog *prog,
enum bpf_attach_type type, u32 flags);
int __cgroup_bpf_detach(struct cgroup *cgrp, struct bpf_prog *prog,
enum bpf_attach_type type, u32 flags);
+int __cgroup_bpf_query(struct cgroup *cgrp, const union bpf_attr *attr,
+ union bpf_attr __user *uattr);
/* Wrapper for __cgroup_bpf_*() protected by cgroup_mutex */
int cgroup_bpf_attach(struct cgroup *cgrp, struct bpf_prog *prog,
enum bpf_attach_type type, u32 flags);
int cgroup_bpf_detach(struct cgroup *cgrp, struct bpf_prog *prog,
enum bpf_attach_type type, u32 flags);
+int cgroup_bpf_query(struct cgroup *cgrp, const union bpf_attr *attr,
+ union bpf_attr __user *uattr);
int __cgroup_bpf_run_filter_skb(struct sock *sk,
struct sk_buff *skb,
diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index a6964b75f070..fc2317f8ea89 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -260,6 +260,9 @@ struct bpf_prog_array {
struct bpf_prog_array __rcu *bpf_prog_array_alloc(u32 prog_cnt, gfp_t flags);
void bpf_prog_array_free(struct bpf_prog_array __rcu *progs);
+int bpf_prog_array_length(struct bpf_prog_array __rcu *progs);
+int bpf_prog_array_copy_to_user(struct bpf_prog_array __rcu *progs,
+ __u32 __user *prog_ids);
#define BPF_PROG_RUN_ARRAY(array, ctx, func) \
({ \
diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index 762f74bc6c47..cb2b9f95160a 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -92,6 +92,7 @@ enum bpf_cmd {
BPF_PROG_GET_FD_BY_ID,
BPF_MAP_GET_FD_BY_ID,
BPF_OBJ_GET_INFO_BY_FD,
+ BPF_PROG_QUERY,
};
enum bpf_map_type {
@@ -211,6 +212,9 @@ enum bpf_attach_type {
/* Specify numa node during map creation */
#define BPF_F_NUMA_NODE (1U << 2)
+/* flags for BPF_PROG_QUERY */
+#define BPF_F_QUERY_EFFECTIVE (1U << 0)
+
#define BPF_OBJ_NAME_LEN 16U
union bpf_attr {
@@ -289,6 +293,15 @@ union bpf_attr {
__u32 info_len;
__aligned_u64 info;
} info;
+
+ struct { /* anonymous struct used by BPF_PROG_QUERY command */
+ __u32 target_fd; /* container object to query */
+ __u32 attach_type;
+ __u32 query_flags;
+ __u32 attach_flags;
+ __aligned_u64 prog_ids;
+ __u32 prog_cnt;
+ } query;
} __attribute__((aligned(8)));
/* BPF helper function descriptions:
diff --git a/kernel/bpf/cgroup.c b/kernel/bpf/cgroup.c
index 70f679a94804..3dd36faa1a99 100644
--- a/kernel/bpf/cgroup.c
+++ b/kernel/bpf/cgroup.c
@@ -384,6 +384,49 @@ int __cgroup_bpf_detach(struct cgroup *cgrp, struct bpf_prog *prog,
return err;
}
+/* Must be called with cgroup_mutex held to avoid races. */
+int __cgroup_bpf_query(struct cgroup *cgrp, const union bpf_attr *attr,
+ union bpf_attr __user *uattr)
+{
+ __u32 __user *prog_ids = u64_to_user_ptr(attr->query.prog_ids);
+ enum bpf_attach_type type = attr->query.attach_type;
+ struct list_head *progs = &cgrp->bpf.progs[type];
+ u32 flags = cgrp->bpf.flags[type];
+ int cnt;
+
+ if (attr->query.query_flags & BPF_F_QUERY_EFFECTIVE)
+ cnt = bpf_prog_array_length(cgrp->bpf.effective[type]);
+ else
+ cnt = prog_list_length(progs);
+
+ if (copy_to_user(&uattr->query.attach_flags, &flags, sizeof(flags)))
+ return -EFAULT;
+ if (copy_to_user(&uattr->query.prog_cnt, &cnt, sizeof(cnt)))
+ return -EFAULT;
+ if (attr->query.prog_cnt == 0 || !prog_ids)
+ /* return early if user requested only program count + flags */
+ return 0;
+ if (attr->query.prog_cnt < cnt)
+ return -ENOSPC;
+
+ if (attr->query.query_flags & BPF_F_QUERY_EFFECTIVE) {
+ return bpf_prog_array_copy_to_user(cgrp->bpf.effective[type],
+ prog_ids);
+ } else {
+ struct bpf_prog_list *pl;
+ u32 id;
+
+ cnt = 0;
+ list_for_each_entry(pl, progs, node) {
+ id = pl->prog->aux->id;
+ if (copy_to_user(prog_ids + cnt, &id, sizeof(id)))
+ return -EFAULT;
+ cnt++;
+ }
+ }
+ return 0;
+}
+
/**
* __cgroup_bpf_run_filter_skb() - Run a program for packet filtering
* @sk: The socket sending or receiving traffic
diff --git a/kernel/bpf/core.c b/kernel/bpf/core.c
index 6b49e1991ae7..6083f15c311a 100644
--- a/kernel/bpf/core.c
+++ b/kernel/bpf/core.c
@@ -1412,6 +1412,39 @@ void bpf_prog_array_free(struct bpf_prog_array __rcu *progs)
kfree_rcu(progs, rcu);
}
+int bpf_prog_array_length(struct bpf_prog_array __rcu *progs)
+{
+ struct bpf_prog **prog;
+ u32 cnt = 0;
+
+ rcu_read_lock();
+ prog = rcu_dereference(progs)->progs;
+ for (; *prog; prog++)
+ cnt++;
+ rcu_read_unlock();
+ return cnt;
+}
+
+int bpf_prog_array_copy_to_user(struct bpf_prog_array __rcu *progs,
+ __u32 __user *prog_ids)
+{
+ struct bpf_prog **prog;
+ u32 cnt = 0, id;
+
+ rcu_read_lock();
+ prog = rcu_dereference(progs)->progs;
+ for (; *prog; prog++) {
+ id = (*prog)->aux->id;
+ if (copy_to_user(prog_ids + cnt, &id, sizeof(id))) {
+ rcu_read_unlock();
+ return -EFAULT;
+ }
+ cnt++;
+ }
+ rcu_read_unlock();
+ return 0;
+}
+
static void bpf_prog_free_deferred(struct work_struct *work)
{
struct bpf_prog_aux *aux;
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index 51bee695d32c..0048cb24ba7b 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -1272,6 +1272,37 @@ static int bpf_prog_detach(const union bpf_attr *attr)
return ret;
}
+#define BPF_PROG_QUERY_LAST_FIELD query.prog_cnt
+
+static int bpf_prog_query(const union bpf_attr *attr,
+ union bpf_attr __user *uattr)
+{
+ struct cgroup *cgrp;
+ int ret;
+
+ if (!capable(CAP_NET_ADMIN))
+ return -EPERM;
+ if (CHECK_ATTR(BPF_PROG_QUERY))
+ return -EINVAL;
+ if (attr->query.query_flags & ~BPF_F_QUERY_EFFECTIVE)
+ return -EINVAL;
+
+ switch (attr->query.attach_type) {
+ case BPF_CGROUP_INET_INGRESS:
+ case BPF_CGROUP_INET_EGRESS:
+ case BPF_CGROUP_INET_SOCK_CREATE:
+ case BPF_CGROUP_SOCK_OPS:
+ break;
+ default:
+ return -EINVAL;
+ }
+ cgrp = cgroup_get_from_fd(attr->query.target_fd);
+ if (IS_ERR(cgrp))
+ return PTR_ERR(cgrp);
+ ret = cgroup_bpf_query(cgrp, attr, uattr);
+ cgroup_put(cgrp);
+ return ret;
+}
#endif /* CONFIG_CGROUP_BPF */
#define BPF_PROG_TEST_RUN_LAST_FIELD test.duration
@@ -1568,6 +1599,9 @@ SYSCALL_DEFINE3(bpf, int, cmd, union bpf_attr __user *, uattr, unsigned int, siz
case BPF_PROG_DETACH:
err = bpf_prog_detach(&attr);
break;
+ case BPF_PROG_QUERY:
+ err = bpf_prog_query(&attr, uattr);
+ break;
#endif
case BPF_PROG_TEST_RUN:
err = bpf_prog_test_run(&attr, uattr);
diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c
index 57eb866ae78d..269512b94a94 100644
--- a/kernel/cgroup/cgroup.c
+++ b/kernel/cgroup/cgroup.c
@@ -5761,4 +5761,14 @@ int cgroup_bpf_detach(struct cgroup *cgrp, struct bpf_prog *prog,
mutex_unlock(&cgroup_mutex);
return ret;
}
+int cgroup_bpf_query(struct cgroup *cgrp, const union bpf_attr *attr,
+ union bpf_attr __user *uattr)
+{
+ int ret;
+
+ mutex_lock(&cgroup_mutex);
+ ret = __cgroup_bpf_query(cgrp, attr, uattr);
+ mutex_unlock(&cgroup_mutex);
+ return ret;
+}
#endif /* CONFIG_CGROUP_BPF */
--
2.9.5
^ permalink raw reply related
* [PATCH net-next 0/8] bpf: muli prog support for cgroup-bpf
From: Alexei Starovoitov @ 2017-10-02 23:48 UTC (permalink / raw)
To: David S . Miller
Cc: Daniel Borkmann, Tejun Heo, David Ahern, netdev, kernel-team
Hi,
cgroup-bpf use cases are getting more advanced and running only
one program per cgroup is no longer enough. Therefore introduce
support for attaching multiple programs per cgroup and running
a set of effective programs.
These patches introduces BPF_F_ALLOW_MULTI flag for BPF_PROG_ATTACH cmd.
The default is still NONE and behavior of BPF_F_ALLOW_OVERRIDE flag
is unchanged.
The difference between three possible flags for BPF_PROG_ATTACH command:
- NONE(default): No further bpf programs allowed in the subtree.
- BPF_F_ALLOW_OVERRIDE: If a sub-cgroup installs some bpf program,
the program in this cgroup yields to sub-cgroup program.
- BPF_F_ALLOW_MULTI: If a sub-cgroup installs some bpf program,
that cgroup program gets run in addition to the program in this cgroup.
Most of the logic is in patch 1. Even when cgroup doesn't have
any programs attached its set of effective program can be non-empty.
To quickly execute them and avoid penalizing cgroups without
any effective programs introduce 'struct bpf_prog_array'
which has an optimization for cgroups with zero effective programs.
Patch 2 introduces BPF_PROG_QUERY command for introspection
Patch 3 makes verifier more strict for cgroup-bpf program types.
Patch 4+ are tests.
More details in individual patches
Alexei Starovoitov (8):
bpf: multi program support for cgroup+bpf
bpf: introduce BPF_PROG_QUERY command
bpf: enforce return code for cgroup-bpf programs
libbpf: introduce bpf_prog_detach2()
samples/bpf: add multi-prog cgroup test case
libbpf: sync bpf.h
libbpf: add support for BPF_PROG_QUERY
samples/bpf: use bpf_prog_query() interface
include/linux/bpf-cgroup.h | 54 ++-
include/linux/bpf.h | 35 ++
include/linux/filter.h | 2 +-
include/uapi/linux/bpf.h | 55 ++-
kernel/bpf/cgroup.c | 510 +++++++++++++++++++++-------
kernel/bpf/core.c | 64 ++++
kernel/bpf/syscall.c | 71 +++-
kernel/bpf/verifier.c | 40 +++
kernel/cgroup/cgroup.c | 38 ++-
samples/bpf/cgroup_helpers.c | 4 +-
samples/bpf/test_cgrp2_attach2.c | 212 +++++++++++-
tools/include/uapi/linux/bpf.h | 55 ++-
tools/lib/bpf/bpf.c | 32 ++
tools/lib/bpf/bpf.h | 4 +-
tools/testing/selftests/bpf/test_verifier.c | 72 ++++
15 files changed, 1066 insertions(+), 182 deletions(-)
--
2.9.5
^ permalink raw reply
* Re: [PATCH] rndis_host: support Novatel Verizon USB730L
From: David Miller @ 2017-10-02 23:17 UTC (permalink / raw)
To: aleksander-Dvg4H30XQSRVIjRurl1/8g
Cc: oliver-GvhC2dPhHPQdnm+yROfE0A, linux-usb-u79uwXL29TY76Z2rM5mHXA,
netdev-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <20170927213103.11987-1-aleksander-Dvg4H30XQSRVIjRurl1/8g@public.gmane.org>
From: Aleksander Morgado <aleksander-Dvg4H30XQSRVIjRurl1/8g@public.gmane.org>
Date: Wed, 27 Sep 2017 23:31:03 +0200
> I'm not sure if binding this logic to a specific vid:pid (1410:9030)
> would be more appropriate here, or if it's ok to just bind
> class/subclass/protocol (as in the activesync case). Let me know
> what you think.
I don't have enough USB Networking knowledge to make a decision here.
Some things seem confusing. For example, if we should be matching
USB_CLASS_MISC, subclass=4, protocol=1 for RNDIS over Ethernet, then
we why aren't we also matching USB_CLASS_MISC, subclass=4, protocol=2
for RNDIS over wireless and instead are matching "Remote RNDIS" in
the form of USB_CLASS_WIRELSS, subclass=1, protocol=3?
I really don't understand any of this magic :-)
So someone more knowledgable needs to review this.
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" 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
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