| Message ID | 20240605-dstats-v1-2-1024396e1670@codeconstruct.com.au (mailing list archive) |
|---|---|
| State | Superseded |
| Delegated to: | Netdev Maintainers |
| Headers | show |
| Series | net: core: Unify dstats with tstats and lstats, add generic collection helper | expand |
Hi Jeremy,
kernel test robot noticed the following build warnings:
[auto build test WARNING on 32f88d65f01bf6f45476d7edbe675e44fb9e1d58]
url: https://github.com/intel-lab-lkp/linux/commits/Jeremy-Kerr/net-core-vrf-Change-pcpu_dstat-fields-to-u64_stats_t/20240605-143942
base: 32f88d65f01bf6f45476d7edbe675e44fb9e1d58
patch link: https://lore.kernel.org/r/20240605-dstats-v1-2-1024396e1670%40codeconstruct.com.au
patch subject: [PATCH 2/3] net: core: Implement dstats-type stats collections
config: openrisc-defconfig (https://download.01.org/0day-ci/archive/20240605/202406051725.14UkSYbV-lkp@intel.com/config)
compiler: or1k-linux-gcc (GCC) 13.2.0
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20240605/202406051725.14UkSYbV-lkp@intel.com/reproduce)
If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <lkp@intel.com>
| Closes: https://lore.kernel.org/oe-kbuild-all/202406051725.14UkSYbV-lkp@intel.com/
All warnings (new ones prefixed by >>):
In file included from ./arch/openrisc/include/generated/asm/rwonce.h:1,
from include/linux/compiler.h:299,
from include/linux/instrumented.h:10,
from include/linux/uaccess.h:6,
from net/core/dev.c:71:
In function '__seqprop_sequence',
inlined from '__u64_stats_fetch_begin' at include/linux/u64_stats_sync.h:171:9,
inlined from 'u64_stats_fetch_begin' at include/linux/u64_stats_sync.h:208:9,
inlined from 'dev_fetch_dstats' at net/core/dev.c:10873:12:
>> include/asm-generic/rwonce.h:44:26: warning: 'dstats' is used uninitialized [-Wuninitialized]
44 | #define __READ_ONCE(x) (*(const volatile __unqual_scalar_typeof(x) *)&(x))
| ~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
include/asm-generic/rwonce.h:50:9: note: in expansion of macro '__READ_ONCE'
50 | __READ_ONCE(x); \
| ^~~~~~~~~~~
include/linux/seqlock.h:211:16: note: in expansion of macro 'READ_ONCE'
211 | return READ_ONCE(s->sequence);
| ^~~~~~~~~
net/core/dev.c: In function 'dev_fetch_dstats':
net/core/dev.c:10868:43: note: 'dstats' was declared here
10868 | const struct pcpu_dstats *dstats;
| ^~~~~~
vim +/dstats +44 include/asm-generic/rwonce.h
e506ea451254ab Will Deacon 2019-10-15 28
e506ea451254ab Will Deacon 2019-10-15 29 /*
e506ea451254ab Will Deacon 2019-10-15 30 * Yes, this permits 64-bit accesses on 32-bit architectures. These will
e506ea451254ab Will Deacon 2019-10-15 31 * actually be atomic in some cases (namely Armv7 + LPAE), but for others we
e506ea451254ab Will Deacon 2019-10-15 32 * rely on the access being split into 2x32-bit accesses for a 32-bit quantity
e506ea451254ab Will Deacon 2019-10-15 33 * (e.g. a virtual address) and a strong prevailing wind.
e506ea451254ab Will Deacon 2019-10-15 34 */
e506ea451254ab Will Deacon 2019-10-15 35 #define compiletime_assert_rwonce_type(t) \
e506ea451254ab Will Deacon 2019-10-15 36 compiletime_assert(__native_word(t) || sizeof(t) == sizeof(long long), \
e506ea451254ab Will Deacon 2019-10-15 37 "Unsupported access size for {READ,WRITE}_ONCE().")
e506ea451254ab Will Deacon 2019-10-15 38
e506ea451254ab Will Deacon 2019-10-15 39 /*
e506ea451254ab Will Deacon 2019-10-15 40 * Use __READ_ONCE() instead of READ_ONCE() if you do not require any
3c9184109e78ea Will Deacon 2019-10-30 41 * atomicity. Note that this may result in tears!
e506ea451254ab Will Deacon 2019-10-15 42 */
b78b331a3f5c07 Will Deacon 2019-10-15 43 #ifndef __READ_ONCE
e506ea451254ab Will Deacon 2019-10-15 @44 #define __READ_ONCE(x) (*(const volatile __unqual_scalar_typeof(x) *)&(x))
b78b331a3f5c07 Will Deacon 2019-10-15 45 #endif
e506ea451254ab Will Deacon 2019-10-15 46
Hi Jeremy,
kernel test robot noticed the following build warnings:
[auto build test WARNING on 32f88d65f01bf6f45476d7edbe675e44fb9e1d58]
url: https://github.com/intel-lab-lkp/linux/commits/Jeremy-Kerr/net-core-vrf-Change-pcpu_dstat-fields-to-u64_stats_t/20240605-143942
base: 32f88d65f01bf6f45476d7edbe675e44fb9e1d58
patch link: https://lore.kernel.org/r/20240605-dstats-v1-2-1024396e1670%40codeconstruct.com.au
patch subject: [PATCH 2/3] net: core: Implement dstats-type stats collections
config: um-allmodconfig (https://download.01.org/0day-ci/archive/20240605/202406051710.KozBBK8o-lkp@intel.com/config)
compiler: clang version 19.0.0git (https://github.com/llvm/llvm-project d7d2d4f53fc79b4b58e8d8d08151b577c3699d4a)
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20240605/202406051710.KozBBK8o-lkp@intel.com/reproduce)
If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <lkp@intel.com>
| Closes: https://lore.kernel.org/oe-kbuild-all/202406051710.KozBBK8o-lkp@intel.com/
All warnings (new ones prefixed by >>):
In file included from net/core/dev.c:80:
In file included from include/linux/sched/isolation.h:5:
In file included from include/linux/cpuset.h:17:
In file included from include/linux/mm.h:2253:
include/linux/vmstat.h:514:36: warning: arithmetic between different enumeration types ('enum node_stat_item' and 'enum lru_list') [-Wenum-enum-conversion]
514 | return node_stat_name(NR_LRU_BASE + lru) + 3; // skip "nr_"
| ~~~~~~~~~~~ ^ ~~~
In file included from net/core/dev.c:80:
In file included from include/linux/sched/isolation.h:7:
In file included from include/linux/tick.h:8:
In file included from include/linux/clockchips.h:14:
In file included from include/linux/clocksource.h:22:
In file included from arch/um/include/asm/io.h:24:
include/asm-generic/io.h:548:31: warning: performing pointer arithmetic on a null pointer has undefined behavior [-Wnull-pointer-arithmetic]
548 | val = __raw_readb(PCI_IOBASE + addr);
| ~~~~~~~~~~ ^
include/asm-generic/io.h:561:61: warning: performing pointer arithmetic on a null pointer has undefined behavior [-Wnull-pointer-arithmetic]
561 | val = __le16_to_cpu((__le16 __force)__raw_readw(PCI_IOBASE + addr));
| ~~~~~~~~~~ ^
include/uapi/linux/byteorder/little_endian.h:37:51: note: expanded from macro '__le16_to_cpu'
37 | #define __le16_to_cpu(x) ((__force __u16)(__le16)(x))
| ^
In file included from net/core/dev.c:80:
In file included from include/linux/sched/isolation.h:7:
In file included from include/linux/tick.h:8:
In file included from include/linux/clockchips.h:14:
In file included from include/linux/clocksource.h:22:
In file included from arch/um/include/asm/io.h:24:
include/asm-generic/io.h:574:61: warning: performing pointer arithmetic on a null pointer has undefined behavior [-Wnull-pointer-arithmetic]
574 | val = __le32_to_cpu((__le32 __force)__raw_readl(PCI_IOBASE + addr));
| ~~~~~~~~~~ ^
include/uapi/linux/byteorder/little_endian.h:35:51: note: expanded from macro '__le32_to_cpu'
35 | #define __le32_to_cpu(x) ((__force __u32)(__le32)(x))
| ^
In file included from net/core/dev.c:80:
In file included from include/linux/sched/isolation.h:7:
In file included from include/linux/tick.h:8:
In file included from include/linux/clockchips.h:14:
In file included from include/linux/clocksource.h:22:
In file included from arch/um/include/asm/io.h:24:
include/asm-generic/io.h:585:33: warning: performing pointer arithmetic on a null pointer has undefined behavior [-Wnull-pointer-arithmetic]
585 | __raw_writeb(value, PCI_IOBASE + addr);
| ~~~~~~~~~~ ^
include/asm-generic/io.h:595:59: warning: performing pointer arithmetic on a null pointer has undefined behavior [-Wnull-pointer-arithmetic]
595 | __raw_writew((u16 __force)cpu_to_le16(value), PCI_IOBASE + addr);
| ~~~~~~~~~~ ^
include/asm-generic/io.h:605:59: warning: performing pointer arithmetic on a null pointer has undefined behavior [-Wnull-pointer-arithmetic]
605 | __raw_writel((u32 __force)cpu_to_le32(value), PCI_IOBASE + addr);
| ~~~~~~~~~~ ^
include/asm-generic/io.h:693:20: warning: performing pointer arithmetic on a null pointer has undefined behavior [-Wnull-pointer-arithmetic]
693 | readsb(PCI_IOBASE + addr, buffer, count);
| ~~~~~~~~~~ ^
include/asm-generic/io.h:701:20: warning: performing pointer arithmetic on a null pointer has undefined behavior [-Wnull-pointer-arithmetic]
701 | readsw(PCI_IOBASE + addr, buffer, count);
| ~~~~~~~~~~ ^
include/asm-generic/io.h:709:20: warning: performing pointer arithmetic on a null pointer has undefined behavior [-Wnull-pointer-arithmetic]
709 | readsl(PCI_IOBASE + addr, buffer, count);
| ~~~~~~~~~~ ^
include/asm-generic/io.h:718:21: warning: performing pointer arithmetic on a null pointer has undefined behavior [-Wnull-pointer-arithmetic]
718 | writesb(PCI_IOBASE + addr, buffer, count);
| ~~~~~~~~~~ ^
include/asm-generic/io.h:727:21: warning: performing pointer arithmetic on a null pointer has undefined behavior [-Wnull-pointer-arithmetic]
727 | writesw(PCI_IOBASE + addr, buffer, count);
| ~~~~~~~~~~ ^
include/asm-generic/io.h:736:21: warning: performing pointer arithmetic on a null pointer has undefined behavior [-Wnull-pointer-arithmetic]
736 | writesl(PCI_IOBASE + addr, buffer, count);
| ~~~~~~~~~~ ^
>> net/core/dev.c:10871:24: warning: variable 'dstats' is uninitialized when used here [-Wuninitialized]
10871 | dstats = per_cpu_ptr(dstats, cpu);
| ^~~~~~
include/linux/percpu-defs.h:263:65: note: expanded from macro 'per_cpu_ptr'
263 | #define per_cpu_ptr(ptr, cpu) ({ (void)(cpu); VERIFY_PERCPU_PTR(ptr); })
| ^~~
include/linux/percpu-defs.h:260:38: note: expanded from macro 'VERIFY_PERCPU_PTR'
260 | (typeof(*(__p)) __kernel __force *)(__p); \
| ^~~
net/core/dev.c:10868:35: note: initialize the variable 'dstats' to silence this warning
10868 | const struct pcpu_dstats *dstats;
| ^
| = NULL
net/core/dev.c:4128:1: warning: unused function 'sch_handle_ingress' [-Wunused-function]
4128 | sch_handle_ingress(struct sk_buff *skb, struct packet_type **pt_prev, int *ret,
| ^~~~~~~~~~~~~~~~~~
net/core/dev.c:4135:1: warning: unused function 'sch_handle_egress' [-Wunused-function]
4135 | sch_handle_egress(struct sk_buff *skb, int *ret, struct net_device *dev)
| ^~~~~~~~~~~~~~~~~
net/core/dev.c:5392:19: warning: unused function 'nf_ingress' [-Wunused-function]
5392 | static inline int nf_ingress(struct sk_buff *skb, struct packet_type **pt_prev,
| ^~~~~~~~~~
17 warnings generated.
vim +/dstats +10871 net/core/dev.c
10851
10852 /**
10853 * dev_fetch_dstats - collate per-cpu network dstats statistics
10854 * @s: place to store stats
10855 * @dstats: per-cpu network stats to read from
10856 *
10857 * Read per-cpu network statistics from dev->dstats and populate the
10858 * related fields in @s.
10859 */
10860 void dev_fetch_dstats(struct rtnl_link_stats64 *s,
10861 const struct pcpu_dstats __percpu *dstats)
10862 {
10863 int cpu;
10864
10865 for_each_possible_cpu(cpu) {
10866 u64 rx_packets, rx_bytes, rx_drops;
10867 u64 tx_packets, tx_bytes, tx_drops;
10868 const struct pcpu_dstats *dstats;
10869 unsigned int start;
10870
10871 dstats = per_cpu_ptr(dstats, cpu);
10872 do {
10873 start = u64_stats_fetch_begin(&dstats->syncp);
10874 rx_packets = u64_stats_read(&dstats->rx_packets);
10875 rx_bytes = u64_stats_read(&dstats->rx_bytes);
10876 rx_drops = u64_stats_read(&dstats->rx_drops);
10877 tx_packets = u64_stats_read(&dstats->tx_packets);
10878 tx_bytes = u64_stats_read(&dstats->tx_bytes);
10879 tx_drops = u64_stats_read(&dstats->tx_drops);
10880 } while (u64_stats_fetch_retry(&dstats->syncp, start));
10881
10882 s->rx_packets += rx_packets;
10883 s->rx_bytes += rx_bytes;
10884 s->rx_dropped += rx_drops;
10885 s->tx_packets += tx_packets;
10886 s->tx_bytes += tx_bytes;
10887 s->tx_dropped += tx_drops;
10888 }
10889 }
10890 EXPORT_SYMBOL_GPL(dev_fetch_dstats);
10891
Hi Jeremy,
kernel test robot noticed the following build warnings:
[auto build test WARNING on 32f88d65f01bf6f45476d7edbe675e44fb9e1d58]
url: https://github.com/intel-lab-lkp/linux/commits/Jeremy-Kerr/net-core-vrf-Change-pcpu_dstat-fields-to-u64_stats_t/20240605-143942
base: 32f88d65f01bf6f45476d7edbe675e44fb9e1d58
patch link: https://lore.kernel.org/r/20240605-dstats-v1-2-1024396e1670%40codeconstruct.com.au
patch subject: [PATCH 2/3] net: core: Implement dstats-type stats collections
config: x86_64-randconfig-121-20240606 (https://download.01.org/0day-ci/archive/20240606/202406060345.LfJ8GAJM-lkp@intel.com/config)
compiler: gcc-13 (Ubuntu 13.2.0-4ubuntu3) 13.2.0
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20240606/202406060345.LfJ8GAJM-lkp@intel.com/reproduce)
If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <lkp@intel.com>
| Closes: https://lore.kernel.org/oe-kbuild-all/202406060345.LfJ8GAJM-lkp@intel.com/
sparse warnings: (new ones prefixed by >>)
net/core/dev.c:3379:23: sparse: sparse: incorrect type in argument 4 (different base types) @@ expected restricted __wsum [usertype] csum @@ got unsigned int @@
net/core/dev.c:3379:23: sparse: expected restricted __wsum [usertype] csum
net/core/dev.c:3379:23: sparse: got unsigned int
net/core/dev.c:3379:23: sparse: sparse: cast from restricted __wsum
>> net/core/dev.c:10871:26: sparse: sparse: incorrect type in initializer (different address spaces) @@ expected void const [noderef] __percpu *__vpp_verify @@ got struct pcpu_dstats const * @@
net/core/dev.c:10871:26: sparse: expected void const [noderef] __percpu *__vpp_verify
net/core/dev.c:10871:26: sparse: got struct pcpu_dstats const *
net/core/dev.c:3819:17: sparse: sparse: context imbalance in '__dev_queue_xmit' - different lock contexts for basic block
net/core/dev.c:5257:17: sparse: sparse: context imbalance in 'net_tx_action' - different lock contexts for basic block
vim +10871 net/core/dev.c
10851
10852 /**
10853 * dev_fetch_dstats - collate per-cpu network dstats statistics
10854 * @s: place to store stats
10855 * @dstats: per-cpu network stats to read from
10856 *
10857 * Read per-cpu network statistics from dev->dstats and populate the
10858 * related fields in @s.
10859 */
10860 void dev_fetch_dstats(struct rtnl_link_stats64 *s,
10861 const struct pcpu_dstats __percpu *dstats)
10862 {
10863 int cpu;
10864
10865 for_each_possible_cpu(cpu) {
10866 u64 rx_packets, rx_bytes, rx_drops;
10867 u64 tx_packets, tx_bytes, tx_drops;
10868 const struct pcpu_dstats *dstats;
10869 unsigned int start;
10870
10871 dstats = per_cpu_ptr(dstats, cpu);
10872 do {
10873 start = u64_stats_fetch_begin(&dstats->syncp);
10874 rx_packets = u64_stats_read(&dstats->rx_packets);
10875 rx_bytes = u64_stats_read(&dstats->rx_bytes);
10876 rx_drops = u64_stats_read(&dstats->rx_drops);
10877 tx_packets = u64_stats_read(&dstats->tx_packets);
10878 tx_bytes = u64_stats_read(&dstats->tx_bytes);
10879 tx_drops = u64_stats_read(&dstats->tx_drops);
10880 } while (u64_stats_fetch_retry(&dstats->syncp, start));
10881
10882 s->rx_packets += rx_packets;
10883 s->rx_bytes += rx_bytes;
10884 s->rx_dropped += rx_drops;
10885 s->tx_packets += tx_packets;
10886 s->tx_bytes += tx_bytes;
10887 s->tx_dropped += tx_drops;
10888 }
10889 }
10890 EXPORT_SYMBOL_GPL(dev_fetch_dstats);
10891
Hi Jeremy,
kernel test robot noticed the following build warnings:
url: https://github.com/intel-lab-lkp/linux/commits/Jeremy-Kerr/net-core-vrf-Change-pcpu_dstat-fields-to-u64_stats_t/20240605-143942
base: 32f88d65f01bf6f45476d7edbe675e44fb9e1d58
patch link: https://lore.kernel.org/r/20240605-dstats-v1-2-1024396e1670%40codeconstruct.com.au
patch subject: [PATCH 2/3] net: core: Implement dstats-type stats collections
config: x86_64-randconfig-161-20240606 (https://download.01.org/0day-ci/archive/20240606/202406061253.ZgaLHWWp-lkp@intel.com/config)
compiler: gcc-13 (Ubuntu 13.2.0-4ubuntu3) 13.2.0
If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <lkp@intel.com>
| Reported-by: Dan Carpenter <dan.carpenter@linaro.org>
| Closes: https://lore.kernel.org/r/202406061253.ZgaLHWWp-lkp@intel.com/
smatch warnings:
net/core/dev.c:10871 dev_fetch_dstats() error: uninitialized symbol 'dstats'.
vim +/dstats +10871 net/core/dev.c
85bc70c64e8362d Jeremy Kerr 2024-06-05 10860 void dev_fetch_dstats(struct rtnl_link_stats64 *s,
85bc70c64e8362d Jeremy Kerr 2024-06-05 10861 const struct pcpu_dstats __percpu *dstats)
^^^^^^
85bc70c64e8362d Jeremy Kerr 2024-06-05 10862 {
85bc70c64e8362d Jeremy Kerr 2024-06-05 10863 int cpu;
85bc70c64e8362d Jeremy Kerr 2024-06-05 10864
85bc70c64e8362d Jeremy Kerr 2024-06-05 10865 for_each_possible_cpu(cpu) {
85bc70c64e8362d Jeremy Kerr 2024-06-05 10866 u64 rx_packets, rx_bytes, rx_drops;
85bc70c64e8362d Jeremy Kerr 2024-06-05 10867 u64 tx_packets, tx_bytes, tx_drops;
85bc70c64e8362d Jeremy Kerr 2024-06-05 10868 const struct pcpu_dstats *dstats;
^^^^^^
Don't declare a local dstats which shadows the function scope variable.
85bc70c64e8362d Jeremy Kerr 2024-06-05 10869 unsigned int start;
85bc70c64e8362d Jeremy Kerr 2024-06-05 10870
85bc70c64e8362d Jeremy Kerr 2024-06-05 @10871 dstats = per_cpu_ptr(dstats, cpu);
85bc70c64e8362d Jeremy Kerr 2024-06-05 10872 do {
85bc70c64e8362d Jeremy Kerr 2024-06-05 10873 start = u64_stats_fetch_begin(&dstats->syncp);
85bc70c64e8362d Jeremy Kerr 2024-06-05 10874 rx_packets = u64_stats_read(&dstats->rx_packets);
85bc70c64e8362d Jeremy Kerr 2024-06-05 10875 rx_bytes = u64_stats_read(&dstats->rx_bytes);
85bc70c64e8362d Jeremy Kerr 2024-06-05 10876 rx_drops = u64_stats_read(&dstats->rx_drops);
85bc70c64e8362d Jeremy Kerr 2024-06-05 10877 tx_packets = u64_stats_read(&dstats->tx_packets);
85bc70c64e8362d Jeremy Kerr 2024-06-05 10878 tx_bytes = u64_stats_read(&dstats->tx_bytes);
85bc70c64e8362d Jeremy Kerr 2024-06-05 10879 tx_drops = u64_stats_read(&dstats->tx_drops);
85bc70c64e8362d Jeremy Kerr 2024-06-05 10880 } while (u64_stats_fetch_retry(&dstats->syncp, start));
85bc70c64e8362d Jeremy Kerr 2024-06-05 10881
85bc70c64e8362d Jeremy Kerr 2024-06-05 10882 s->rx_packets += rx_packets;
85bc70c64e8362d Jeremy Kerr 2024-06-05 10883 s->rx_bytes += rx_bytes;
85bc70c64e8362d Jeremy Kerr 2024-06-05 10884 s->rx_dropped += rx_drops;
85bc70c64e8362d Jeremy Kerr 2024-06-05 10885 s->tx_packets += tx_packets;
85bc70c64e8362d Jeremy Kerr 2024-06-05 10886 s->tx_bytes += tx_bytes;
85bc70c64e8362d Jeremy Kerr 2024-06-05 10887 s->tx_dropped += tx_drops;
85bc70c64e8362d Jeremy Kerr 2024-06-05 10888 }
85bc70c64e8362d Jeremy Kerr 2024-06-05 10889 }
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index f148a01dd1d1..fdc3d8a6c0f4 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -4690,6 +4690,9 @@ void netdev_stats_to_stats64(struct rtnl_link_stats64 *stats64, void dev_fetch_sw_netstats(struct rtnl_link_stats64 *s, const struct pcpu_sw_netstats __percpu *netstats); void dev_get_tstats64(struct net_device *dev, struct rtnl_link_stats64 *s); +void dev_fetch_dstats(struct rtnl_link_stats64 *s, + const struct pcpu_dstats __percpu *dstats); +void dev_get_dstats64(struct net_device *dev, struct rtnl_link_stats64 *s); enum { NESTED_SYNC_IMM_BIT, diff --git a/net/core/dev.c b/net/core/dev.c index e1bb6d7856d9..dc77529600ce 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -10849,6 +10849,62 @@ void dev_get_tstats64(struct net_device *dev, struct rtnl_link_stats64 *s) } EXPORT_SYMBOL_GPL(dev_get_tstats64); +/** + * dev_fetch_dstats - collate per-cpu network dstats statistics + * @s: place to store stats + * @dstats: per-cpu network stats to read from + * + * Read per-cpu network statistics from dev->dstats and populate the + * related fields in @s. + */ +void dev_fetch_dstats(struct rtnl_link_stats64 *s, + const struct pcpu_dstats __percpu *dstats) +{ + int cpu; + + for_each_possible_cpu(cpu) { + u64 rx_packets, rx_bytes, rx_drops; + u64 tx_packets, tx_bytes, tx_drops; + const struct pcpu_dstats *dstats; + unsigned int start; + + dstats = per_cpu_ptr(dstats, cpu); + do { + start = u64_stats_fetch_begin(&dstats->syncp); + rx_packets = u64_stats_read(&dstats->rx_packets); + rx_bytes = u64_stats_read(&dstats->rx_bytes); + rx_drops = u64_stats_read(&dstats->rx_drops); + tx_packets = u64_stats_read(&dstats->tx_packets); + tx_bytes = u64_stats_read(&dstats->tx_bytes); + tx_drops = u64_stats_read(&dstats->tx_drops); + } while (u64_stats_fetch_retry(&dstats->syncp, start)); + + s->rx_packets += rx_packets; + s->rx_bytes += rx_bytes; + s->rx_dropped += rx_drops; + s->tx_packets += tx_packets; + s->tx_bytes += tx_bytes; + s->tx_dropped += tx_drops; + } +} +EXPORT_SYMBOL_GPL(dev_fetch_dstats); + +/** + * dev_get_dstats64 - ndo_get_stats64 implementation for dtstats-based + * account. + * @dev: device to get statistics from + * @s: place to store stats + * + * Populate @s from dev->stats and dev->dstats. Can be used as + * ndo_get_stats64() callback. + */ +void dev_get_dstats64(struct net_device *dev, struct rtnl_link_stats64 *s) +{ + netdev_stats_to_stats64(s, &dev->stats); + dev_fetch_dstats(s, dev->dstats); +} +EXPORT_SYMBOL_GPL(dev_get_dstats64); + struct netdev_queue *dev_ingress_queue_create(struct net_device *dev) { struct netdev_queue *queue = dev_ingress_queue(dev);
We currently have dev_get_tstats64() for collecting per-cpu stats of type pcpu_sw_netstats ("tstats"). However, tstats doesn't allow for accounting tx/rx drops. We do have a stats variant that does have stats for dropped packets: struct pcpu_dstats, but there are no core helpers for using those stats. The VRF driver uses dstats, by providing its own collation/fetch functions to do so. This change adds a common helpers for dstats-type collection, based on the VRF driver's own (plus the unused tx_drops stat from there). We will switch the VRF driver to use this in the next change. Signed-off-by: Jeremy Kerr <jk@codeconstruct.com.au> --- include/linux/netdevice.h | 3 +++ net/core/dev.c | 56 +++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 59 insertions(+)