* [PATCH v3 bpf-next 03/14] samples: bpf: makefile: use --target from cross-compile
From: Ivan Khoronzhuk @ 2019-09-16 10:54 UTC (permalink / raw)
To: ast, daniel, yhs, davem, jakub.kicinski, hawk, john.fastabend
Cc: linux-kernel, netdev, bpf, clang-built-linux, sergei.shtylyov,
Ivan Khoronzhuk
In-Reply-To: <20190916105433.11404-1-ivan.khoronzhuk@linaro.org>
For cross compiling the target triple can be inherited from
cross-compile prefix as it's done in CLANG_FLAGS from kernel makefile.
So copy-paste this decision from kernel Makefile.
Signed-off-by: Ivan Khoronzhuk <ivan.khoronzhuk@linaro.org>
---
samples/bpf/Makefile | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/samples/bpf/Makefile b/samples/bpf/Makefile
index 43dee90dffa4..b59e77e2250e 100644
--- a/samples/bpf/Makefile
+++ b/samples/bpf/Makefile
@@ -195,7 +195,7 @@ BTF_PAHOLE ?= pahole
# Detect that we're cross compiling and use the cross compiler
ifdef CROSS_COMPILE
HOSTCC = $(CROSS_COMPILE)gcc
-CLANG_ARCH_ARGS = -target $(ARCH)
+CLANG_ARCH_ARGS = --target=$(notdir $(CROSS_COMPILE:%-=%))
endif
# Don't evaluate probes and warnings if we need to run make recursively
--
2.17.1
^ permalink raw reply related
* [PATCH v3 bpf-next 01/14] samples: bpf: makefile: fix HDR_PROBE "echo"
From: Ivan Khoronzhuk @ 2019-09-16 10:54 UTC (permalink / raw)
To: ast, daniel, yhs, davem, jakub.kicinski, hawk, john.fastabend
Cc: linux-kernel, netdev, bpf, clang-built-linux, sergei.shtylyov,
Ivan Khoronzhuk
In-Reply-To: <20190916105433.11404-1-ivan.khoronzhuk@linaro.org>
echo should be replaced with echo -e to handle '\n' correctly, but
instead, replace it with printf as some systems can't handle echo -e.
Signed-off-by: Ivan Khoronzhuk <ivan.khoronzhuk@linaro.org>
---
samples/bpf/Makefile | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/samples/bpf/Makefile b/samples/bpf/Makefile
index 1d9be26b4edd..f50ca852c2a8 100644
--- a/samples/bpf/Makefile
+++ b/samples/bpf/Makefile
@@ -201,7 +201,7 @@ endif
# Don't evaluate probes and warnings if we need to run make recursively
ifneq ($(src),)
-HDR_PROBE := $(shell echo "\#include <linux/types.h>\n struct list_head { int a; }; int main() { return 0; }" | \
+HDR_PROBE := $(shell printf "\#include <linux/types.h>\n struct list_head { int a; }; int main() { return 0; }" | \
$(HOSTCC) $(KBUILD_HOSTCFLAGS) -x c - -o /dev/null 2>/dev/null && \
echo okay)
--
2.17.1
^ permalink raw reply related
* [PATCH v3 bpf-next 00/14] samples: bpf: improve/fix cross-compilation
From: Ivan Khoronzhuk @ 2019-09-16 10:54 UTC (permalink / raw)
To: ast, daniel, yhs, davem, jakub.kicinski, hawk, john.fastabend
Cc: linux-kernel, netdev, bpf, clang-built-linux, sergei.shtylyov,
Ivan Khoronzhuk
This series contains mainly fixes/improvements for cross-compilation
but not only, tested for arm, arm64, but intended for any arch.
Also verified on native build (not cross compilation) for x86_64
and arm.
Initial RFC link:
https://lkml.org/lkml/2019/8/29/1665
Prev. version:
https://lkml.org/lkml/2019/9/10/331
Besides the patches given here, the RFC also contains couple patches
related to llvm clang
arm: include: asm: swab: mask rev16 instruction for clang
arm: include: asm: unified: mask .syntax unified for clang
They are necessarily to verify arm build.
The change touches not only cross-compilation and can have impact on
other archs and build environments, so might be good idea to verify
it in order to add appropriate changes, some warn options could be
tuned also.
All is tested on x86-64 with clang installed (has to be built containing
targets for arm, arm64..., see llc --version, usually it's present already)
Instructions to test native on x86_64
=================================================
Native build on x86_64 is done in usual way and shouldn't have difference
except HOSTCC is now printed as CC wile building the samples.
Instructions to test cross compilation on arm64
=================================================
#Toolchain used for test:
gcc version 8.3.0
(GNU Toolchain for the A-profile Architecture 8.3-2019.03 (arm-rel-8.36))
# Get some arm64 FS, containing at least libelf
I've used sdk for TI am65x got here:
http://downloads.ti.com/processor-sdk-linux/esd/AM65X/latest/exports/\
ti-processor-sdk-linux-am65xx-evm-06.00.00.07-Linux-x86-Install.bin
# Install this binary to some dir, say "sdk".
# Configure kernel (use defconfig as no matter), but clean everything
# before.
make ARCH=arm64 -C tools/ clean
make ARCH=arm64 -C samples/bpf clean
make ARCH=arm64 clean
make ARCH=arm64 defconfig
# The kernel version used in sdk doesn't correspond to checked one,
# but for this verification only headers need to be syched,
# so install them:
make ARCH=arm64 headers_install
# or on SDK if need keep them in sync (not necessarily to verify):
make ARCH=arm64 INSTALL_HDR_PATH=/../sdk/\
ti-processor-sdk-linux-am65xx-evm-06.00.00.07/linux-devkit/sysroots/\
aarch64-linux/usr headers_install
# Build samples
make samples/bpf/ ARCH=arm64 CROSS_COMPILE="aarch64-linux-gnu-"\
SYSROOT="/../sdk/ti-processor-sdk-linux-am65xx-evm-06.00.00.07/\
linux-devkit/sysroots/aarch64-linux"
Instructions to test cross compilation on arm
=================================================
#Toolchains used for test:
arm-linux-gnueabihf-gcc (Linaro GCC 7.2-2017.11) 7.2.1 20171011
or
arm-linux-gnueabihf-gcc
(GNU Toolchain for the A-profile Architecture 8.3-2019.03 \
(arm-rel-8.36)) 8.3.0
# Get some FS, I've used sdk for TI am52xx got here:
http://downloads.ti.com/processor-sdk-linux/esd/AM57X/05_03_00_07/exports/\
ti-processor-sdk-linux-am57xx-evm-05.03.00.07-Linux-x86-Install.bin
# Install this binary to some dir, say "sdk".
# Configure kernel, but clean everything before.
make ARCH=arm -C tools/ clean
make ARCH=arm -C samples/bpf clean
make ARCH=arm clean
make ARCH=arm omap2plus_defconfig
# The kernel version used in sdk doesn't correspond to checked one, but
headers only should be synched, so install them:
make ARCH=arm64 headers_install
# or on SDK if need keep them in sync (not necessarily):
make ARCH=arm INSTALL_HDR_PATH=/../sdk/\
ti-processor-sdk-linux-am57xx-evm-05.03.00.07/linux-devkit/sysroots/\
armv7ahf-neon-linux-gnueabi/usr headers_install
# Build samples
make samples/bpf/ ARCH=arm CROSS_COMPILE="arm-linux-gnueabihf-"\
SYSROOT="/../sdk/ti-processor-sdk-linux-am57xx-evm-05.03\
.00.07/linux-devkit/sysroots/armv7ahf-neon-linux-gnueabi"
Based on bpf-next/master
v3..v2:
- renamed makefile.progs to makeifle.target, as more appropriate
- left only __LINUX_ARM_ARCH__ for D options for arm
- for host build - left options from KBUILD_HOST for compatibility reasons
- split patch adding c/cxx/ld flags to libbpf by modules
- moved readme change to separate patch
- added patch setting options for cross-compile
- fixed issue with option error for syscall_nrs.S,
avoiding overlap for ccflags-y.
v2..v1:
- restructured patches order
- split "samples: bpf: Makefile: base progs build on Makefile.progs"
to make change more readable. It added couple nice extra patches.
- removed redundant patch:
"samples: bpf: Makefile: remove target for native build"
- added fix:
"samples: bpf: makefile: fix cookie_uid_helper_example obj build"
- limited -D option filter only for arm
- improved comments
- added couple instructions to verify cross compilation for arm and
arm64 arches based on TI am57xx and am65xx sdks.
- corrected include a little order
Ivan Khoronzhuk (14):
samples: bpf: makefile: fix HDR_PROBE "echo"
samples: bpf: makefile: fix cookie_uid_helper_example obj build
samples: bpf: makefile: use --target from cross-compile
samples: bpf: use own EXTRA_CFLAGS for clang commands
samples: bpf: makefile: use __LINUX_ARM_ARCH__ selector for arm
samples: bpf: makefile: drop unnecessarily inclusion for bpf_load
samples: bpf: add makefile.target for separate CC target build
samples: bpf: makefile: base target programs rules on Makefile.target
samples: bpf: makefile: use own flags but not host when cross compile
samples: bpf: makefile: use target CC environment for HDR_PROBE
libbpf: makefile: add C/CXX/LDFLAGS to libbpf.so and test_libpf
targets
samples: bpf: makefile: provide C/CXX/LD flags to libbpf
samples: bpf: makefile: add sysroot support
samples: bpf: README: add preparation steps and sysroot info
samples/bpf/Makefile | 179 +++++++++++++++++++++---------------
samples/bpf/Makefile.target | 75 +++++++++++++++
samples/bpf/README.rst | 41 ++++++++-
tools/lib/bpf/Makefile | 11 ++-
4 files changed, 225 insertions(+), 81 deletions(-)
create mode 100644 samples/bpf/Makefile.target
--
2.17.1
^ permalink raw reply
* pull-request: bpf-next 2019-09-16
From: Daniel Borkmann @ 2019-09-16 10:26 UTC (permalink / raw)
To: davem; +Cc: daniel, ast, netdev, bpf
Hi David,
The following pull-request contains BPF updates for your *net-next* tree.
The main changes are:
1) Now that initial BPF backend for gcc has been merged upstream, enable
BPF kselftest suite for bpf-gcc. Also fix a BE issue with access to
bpf_sysctl.file_pos, from Ilya.
2) Follow-up fix for link-vmlinux.sh to remove bash-specific extensions
related to recent work on exposing BTF info through sysfs, from Andrii.
3) AF_XDP zero copy fixes for i40e and ixgbe driver which caused umem
headroom to be added twice, from Ciara.
4) Refactoring work to convert sock opt tests into test_progs framework
in BPF kselftests, from Stanislav.
5) Fix a general protection fault in dev_map_hash_update_elem(), from Toke.
6) Cleanup to use BPF_PROG_RUN() macro in KCM, from Sami.
Please consider pulling these changes from:
git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next.git
Thanks a lot!
----------------------------------------------------------------
The following changes since commit 1e46c09ec10049a9e366153b32e41cc557383fdb:
Merge git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next (2019-09-06 16:49:17 +0200)
are available in the Git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next.git
for you to fetch changes up to d895a0f16fadb26d22ab531c49768f7642ae5c3e:
bpf: fix accessing bpf_sysctl.file_pos on s390 (2019-09-16 11:44:05 +0200)
----------------------------------------------------------------
Alexei Starovoitov (1):
Merge branch 'move-sockopt-tests'
Andrii Nakryiko (1):
kbuild: replace BASH-specific ${@:2} with shift and ${@}
Ciara Loftus (3):
i40e: fix xdp handle calculations
ixgbe: fix xdp handle calculations
samples/bpf: fix xdpsock l2fwd tx for unaligned mode
Daniel Borkmann (1):
Merge branch 'bpf-af-xdp-unaligned-fixes'
Ilya Leoshkevich (2):
selftests/bpf: add bpf-gcc support
bpf: fix accessing bpf_sysctl.file_pos on s390
Sami Tolvanen (1):
kcm: use BPF_PROG_RUN
Stanislav Fomichev (6):
selftests/bpf: test_progs: add test__join_cgroup helper
selftests/bpf: test_progs: convert test_sockopt
selftests/bpf: test_progs: convert test_sockopt_sk
selftests/bpf: test_progs: convert test_sockopt_multi
selftests/bpf: test_progs: convert test_sockopt_inherit
selftests/bpf: test_progs: convert test_tcp_rtt
Toke Høiland-Jørgensen (1):
xdp: Fix race in dev_map_hash_update_elem() when replacing element
drivers/net/ethernet/intel/i40e/i40e_xsk.c | 4 +-
drivers/net/ethernet/intel/ixgbe/ixgbe_xsk.c | 4 +-
include/linux/filter.h | 8 +-
kernel/bpf/cgroup.c | 10 +-
kernel/bpf/devmap.c | 17 +++-
kernel/bpf/verifier.c | 4 +-
net/kcm/kcmsock.c | 2 +-
samples/bpf/xdpsock_user.c | 2 +-
scripts/link-vmlinux.sh | 16 +++-
tools/testing/selftests/bpf/.gitignore | 5 -
tools/testing/selftests/bpf/Makefile | 77 +++++++++++-----
tools/testing/selftests/bpf/bpf_helpers.h | 24 +++--
.../bpf/{test_sockopt.c => prog_tests/sockopt.c} | 50 ++--------
.../sockopt_inherit.c} | 102 +++++++++------------
.../sockopt_multi.c} | 62 ++-----------
.../{test_sockopt_sk.c => prog_tests/sockopt_sk.c} | 60 +++---------
.../bpf/{test_tcp_rtt.c => prog_tests/tcp_rtt.c} | 83 ++++++-----------
tools/testing/selftests/bpf/progs/test_tc_edt.c | 1 +
tools/testing/selftests/bpf/test_progs.c | 38 ++++++++
tools/testing/selftests/bpf/test_progs.h | 4 +-
tools/testing/selftests/bpf/test_sysctl.c | 9 +-
21 files changed, 260 insertions(+), 322 deletions(-)
rename tools/testing/selftests/bpf/{test_sockopt.c => prog_tests/sockopt.c} (96%)
rename tools/testing/selftests/bpf/{test_sockopt_inherit.c => prog_tests/sockopt_inherit.c} (72%)
rename tools/testing/selftests/bpf/{test_sockopt_multi.c => prog_tests/sockopt_multi.c} (83%)
rename tools/testing/selftests/bpf/{test_sockopt_sk.c => prog_tests/sockopt_sk.c} (79%)
rename tools/testing/selftests/bpf/{test_tcp_rtt.c => prog_tests/tcp_rtt.c} (76%)
^ permalink raw reply
* [PATCH V2 net-next 01/11] net: ena: add intr_moder_rx_interval to struct ena_com_dev and use it
From: akiyano @ 2019-09-16 11:31 UTC (permalink / raw)
To: davem, netdev
Cc: Arthur Kiyanovski, dwmw, zorik, matua, saeedb, msw, aliguori,
nafea, gtzalik, netanel, alisaidi, benh, sameehj, ndagan
In-Reply-To: <1568633496-4143-1-git-send-email-akiyano@amazon.com>
From: Arthur Kiyanovski <akiyano@amazon.com>
Add intr_moder_rx_interval to struct ena_com_dev and use it as the
location where the interrupt moderation rx interval is saved, instead
of the interrupt moderation table.
This is done as a first step before removing the old interrupt moderation
code.
Signed-off-by: Arthur Kiyanovski <akiyano@amazon.com>
---
drivers/net/ethernet/amazon/ena/ena_com.c | 20 ++++----------------
drivers/net/ethernet/amazon/ena/ena_com.h | 8 +++++++-
drivers/net/ethernet/amazon/ena/ena_netdev.c | 3 ++-
3 files changed, 13 insertions(+), 18 deletions(-)
diff --git a/drivers/net/ethernet/amazon/ena/ena_com.c b/drivers/net/ethernet/amazon/ena/ena_com.c
index 911a2e7a375a..4144936bf8a2 100644
--- a/drivers/net/ethernet/amazon/ena/ena_com.c
+++ b/drivers/net/ethernet/amazon/ena/ena_com.c
@@ -1297,9 +1297,6 @@ static int ena_com_init_interrupt_moderation_table(struct ena_com_dev *ena_dev)
static void ena_com_update_intr_delay_resolution(struct ena_com_dev *ena_dev,
u16 intr_delay_resolution)
{
- struct ena_intr_moder_entry *intr_moder_tbl = ena_dev->intr_moder_tbl;
- unsigned int i;
-
if (!intr_delay_resolution) {
pr_err("Illegal intr_delay_resolution provided. Going to use default 1 usec resolution\n");
intr_delay_resolution = 1;
@@ -1307,8 +1304,7 @@ static void ena_com_update_intr_delay_resolution(struct ena_com_dev *ena_dev,
ena_dev->intr_delay_resolution = intr_delay_resolution;
/* update Rx */
- for (i = 0; i < ENA_INTR_MAX_NUM_OF_LEVELS; i++)
- intr_moder_tbl[i].intr_moder_interval /= intr_delay_resolution;
+ ena_dev->intr_moder_rx_interval /= intr_delay_resolution;
/* update Tx */
ena_dev->intr_moder_tx_interval /= intr_delay_resolution;
@@ -2798,11 +2794,8 @@ int ena_com_update_nonadaptive_moderation_interval_rx(struct ena_com_dev *ena_de
return -EFAULT;
}
- /* We use LOWEST entry of moderation table for storing
- * nonadaptive interrupt coalescing values
- */
- ena_dev->intr_moder_tbl[ENA_INTR_MODER_LOWEST].intr_moder_interval =
- rx_coalesce_usecs / ena_dev->intr_delay_resolution;
+ ena_dev->intr_moder_rx_interval = rx_coalesce_usecs /
+ ena_dev->intr_delay_resolution;
return 0;
}
@@ -2907,12 +2900,7 @@ unsigned int ena_com_get_nonadaptive_moderation_interval_tx(struct ena_com_dev *
unsigned int ena_com_get_nonadaptive_moderation_interval_rx(struct ena_com_dev *ena_dev)
{
- struct ena_intr_moder_entry *intr_moder_tbl = ena_dev->intr_moder_tbl;
-
- if (intr_moder_tbl)
- return intr_moder_tbl[ENA_INTR_MODER_LOWEST].intr_moder_interval;
-
- return 0;
+ return ena_dev->intr_moder_rx_interval;
}
void ena_com_init_intr_moderation_entry(struct ena_com_dev *ena_dev,
diff --git a/drivers/net/ethernet/amazon/ena/ena_com.h b/drivers/net/ethernet/amazon/ena/ena_com.h
index 0d3664fe260d..baeefc6af4f3 100644
--- a/drivers/net/ethernet/amazon/ena/ena_com.h
+++ b/drivers/net/ethernet/amazon/ena/ena_com.h
@@ -93,7 +93,7 @@
#define ENA_INTR_HIGHEST_BYTES (192 * 1024)
#define ENA_INTR_INITIAL_TX_INTERVAL_USECS 196
-#define ENA_INTR_INITIAL_RX_INTERVAL_USECS 4
+#define ENA_INTR_INITIAL_RX_INTERVAL_USECS 0
#define ENA_INTR_DELAY_OLD_VALUE_WEIGHT 6
#define ENA_INTR_DELAY_NEW_VALUE_WEIGHT 4
#define ENA_INTR_MODER_LEVEL_STRIDE 2
@@ -376,7 +376,13 @@ struct ena_com_dev {
struct ena_host_attribute host_attr;
bool adaptive_coalescing;
u16 intr_delay_resolution;
+
+ /* interrupt moderation intervals are in usec divided by
+ * intr_delay_resolution, which is supplied by the device.
+ */
u32 intr_moder_tx_interval;
+ u32 intr_moder_rx_interval;
+
struct ena_intr_moder_entry *intr_moder_tbl;
struct ena_com_llq_info llq_info;
diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.c b/drivers/net/ethernet/amazon/ena/ena_netdev.c
index 664e3ed97ea9..233b252ceb9e 100644
--- a/drivers/net/ethernet/amazon/ena/ena_netdev.c
+++ b/drivers/net/ethernet/amazon/ena/ena_netdev.c
@@ -3485,10 +3485,11 @@ static int ena_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
calc_queue_ctx.get_feat_ctx = &get_feat_ctx;
calc_queue_ctx.pdev = pdev;
- /* initial Tx interrupt delay, Assumes 1 usec granularity.
+ /* Initial Tx and RX interrupt delay. Assumes 1 usec granularity.
* Updated during device initialization with the real granularity
*/
ena_dev->intr_moder_tx_interval = ENA_INTR_INITIAL_TX_INTERVAL_USECS;
+ ena_dev->intr_moder_rx_interval = ENA_INTR_INITIAL_RX_INTERVAL_USECS;
io_queue_num = ena_calc_io_queue_num(pdev, ena_dev, &get_feat_ctx);
rc = ena_calc_queue_size(&calc_queue_ctx);
if (rc || io_queue_num <= 0) {
--
2.17.2
^ permalink raw reply related
* [PATCH V2 net-next 03/11] net: ena: reimplement set/get_coalesce()
From: akiyano @ 2019-09-16 11:31 UTC (permalink / raw)
To: davem, netdev
Cc: Arthur Kiyanovski, dwmw, zorik, matua, saeedb, msw, aliguori,
nafea, gtzalik, netanel, alisaidi, benh, sameehj, ndagan
In-Reply-To: <1568633496-4143-1-git-send-email-akiyano@amazon.com>
From: Arthur Kiyanovski <akiyano@amazon.com>
1. Remove old adaptive interrupt moderation code from set/get_coalesce()
2. Add ena_update_rx_rings_intr_moderation() function for updating
nonadaptive interrupt moderation intervals similarly to
ena_update_tx_rings_intr_moderation().
3. Remove checks of multiple unsupported received interrupt coalescing
parameters. This makes code cleaner and cancels the need to update
it every time a new coalescing parameter is invented.
Signed-off-by: Arthur Kiyanovski <akiyano@amazon.com>
---
drivers/net/ethernet/amazon/ena/ena_ethtool.c | 84 ++++++-------------
1 file changed, 26 insertions(+), 58 deletions(-)
diff --git a/drivers/net/ethernet/amazon/ena/ena_ethtool.c b/drivers/net/ethernet/amazon/ena/ena_ethtool.c
index b997c3ce9e2b..0f90e2296630 100644
--- a/drivers/net/ethernet/amazon/ena/ena_ethtool.c
+++ b/drivers/net/ethernet/amazon/ena/ena_ethtool.c
@@ -305,7 +305,6 @@ static int ena_get_coalesce(struct net_device *net_dev,
{
struct ena_adapter *adapter = netdev_priv(net_dev);
struct ena_com_dev *ena_dev = adapter->ena_dev;
- struct ena_intr_moder_entry intr_moder_entry;
if (!ena_com_interrupt_moderation_supported(ena_dev)) {
/* the devie doesn't support interrupt moderation */
@@ -314,23 +313,12 @@ static int ena_get_coalesce(struct net_device *net_dev,
coalesce->tx_coalesce_usecs =
ena_com_get_nonadaptive_moderation_interval_tx(ena_dev) /
ena_dev->intr_delay_resolution;
- if (!ena_com_get_adaptive_moderation_enabled(ena_dev)) {
+
+ if (!ena_com_get_adaptive_moderation_enabled(ena_dev))
coalesce->rx_coalesce_usecs =
ena_com_get_nonadaptive_moderation_interval_rx(ena_dev)
/ ena_dev->intr_delay_resolution;
- } else {
- ena_com_get_intr_moderation_entry(adapter->ena_dev, ENA_INTR_MODER_LOWEST, &intr_moder_entry);
- coalesce->rx_coalesce_usecs_low = intr_moder_entry.intr_moder_interval;
- coalesce->rx_max_coalesced_frames_low = intr_moder_entry.pkts_per_interval;
-
- ena_com_get_intr_moderation_entry(adapter->ena_dev, ENA_INTR_MODER_MID, &intr_moder_entry);
- coalesce->rx_coalesce_usecs = intr_moder_entry.intr_moder_interval;
- coalesce->rx_max_coalesced_frames = intr_moder_entry.pkts_per_interval;
-
- ena_com_get_intr_moderation_entry(adapter->ena_dev, ENA_INTR_MODER_HIGHEST, &intr_moder_entry);
- coalesce->rx_coalesce_usecs_high = intr_moder_entry.intr_moder_interval;
- coalesce->rx_max_coalesced_frames_high = intr_moder_entry.pkts_per_interval;
- }
+
coalesce->use_adaptive_rx_coalesce =
ena_com_get_adaptive_moderation_enabled(ena_dev);
@@ -348,12 +336,22 @@ static void ena_update_tx_rings_intr_moderation(struct ena_adapter *adapter)
adapter->tx_ring[i].smoothed_interval = val;
}
+static void ena_update_rx_rings_intr_moderation(struct ena_adapter *adapter)
+{
+ unsigned int val;
+ int i;
+
+ val = ena_com_get_nonadaptive_moderation_interval_rx(adapter->ena_dev);
+
+ for (i = 0; i < adapter->num_queues; i++)
+ adapter->rx_ring[i].smoothed_interval = val;
+}
+
static int ena_set_coalesce(struct net_device *net_dev,
struct ethtool_coalesce *coalesce)
{
struct ena_adapter *adapter = netdev_priv(net_dev);
struct ena_com_dev *ena_dev = adapter->ena_dev;
- struct ena_intr_moder_entry intr_moder_entry;
int rc;
if (!ena_com_interrupt_moderation_supported(ena_dev)) {
@@ -361,22 +359,6 @@ static int ena_set_coalesce(struct net_device *net_dev,
return -EOPNOTSUPP;
}
- if (coalesce->rx_coalesce_usecs_irq ||
- coalesce->rx_max_coalesced_frames_irq ||
- coalesce->tx_coalesce_usecs_irq ||
- coalesce->tx_max_coalesced_frames ||
- coalesce->tx_max_coalesced_frames_irq ||
- coalesce->stats_block_coalesce_usecs ||
- coalesce->use_adaptive_tx_coalesce ||
- coalesce->pkt_rate_low ||
- coalesce->tx_coalesce_usecs_low ||
- coalesce->tx_max_coalesced_frames_low ||
- coalesce->pkt_rate_high ||
- coalesce->tx_coalesce_usecs_high ||
- coalesce->tx_max_coalesced_frames_high ||
- coalesce->rate_sample_interval)
- return -EINVAL;
-
rc = ena_com_update_nonadaptive_moderation_interval_tx(ena_dev,
coalesce->tx_coalesce_usecs);
if (rc)
@@ -384,37 +366,23 @@ static int ena_set_coalesce(struct net_device *net_dev,
ena_update_tx_rings_intr_moderation(adapter);
- if (ena_com_get_adaptive_moderation_enabled(ena_dev)) {
- if (!coalesce->use_adaptive_rx_coalesce) {
- ena_com_disable_adaptive_moderation(ena_dev);
- rc = ena_com_update_nonadaptive_moderation_interval_rx(ena_dev,
- coalesce->rx_coalesce_usecs);
- return rc;
- }
- } else { /* was in non-adaptive mode */
- if (coalesce->use_adaptive_rx_coalesce) {
+ if (coalesce->use_adaptive_rx_coalesce) {
+ if (!ena_com_get_adaptive_moderation_enabled(ena_dev))
ena_com_enable_adaptive_moderation(ena_dev);
- } else {
- rc = ena_com_update_nonadaptive_moderation_interval_rx(ena_dev,
- coalesce->rx_coalesce_usecs);
- return rc;
- }
+ return 0;
}
- intr_moder_entry.intr_moder_interval = coalesce->rx_coalesce_usecs_low;
- intr_moder_entry.pkts_per_interval = coalesce->rx_max_coalesced_frames_low;
- intr_moder_entry.bytes_per_interval = ENA_INTR_BYTE_COUNT_NOT_SUPPORTED;
- ena_com_init_intr_moderation_entry(adapter->ena_dev, ENA_INTR_MODER_LOWEST, &intr_moder_entry);
+ rc = ena_com_update_nonadaptive_moderation_interval_rx(ena_dev,
+ coalesce->rx_coalesce_usecs);
+ if (rc)
+ return rc;
- intr_moder_entry.intr_moder_interval = coalesce->rx_coalesce_usecs;
- intr_moder_entry.pkts_per_interval = coalesce->rx_max_coalesced_frames;
- intr_moder_entry.bytes_per_interval = ENA_INTR_BYTE_COUNT_NOT_SUPPORTED;
- ena_com_init_intr_moderation_entry(adapter->ena_dev, ENA_INTR_MODER_MID, &intr_moder_entry);
+ ena_update_rx_rings_intr_moderation(adapter);
- intr_moder_entry.intr_moder_interval = coalesce->rx_coalesce_usecs_high;
- intr_moder_entry.pkts_per_interval = coalesce->rx_max_coalesced_frames_high;
- intr_moder_entry.bytes_per_interval = ENA_INTR_BYTE_COUNT_NOT_SUPPORTED;
- ena_com_init_intr_moderation_entry(adapter->ena_dev, ENA_INTR_MODER_HIGHEST, &intr_moder_entry);
+ if (!coalesce->use_adaptive_rx_coalesce) {
+ if (ena_com_get_adaptive_moderation_enabled(ena_dev))
+ ena_com_disable_adaptive_moderation(ena_dev);
+ }
return 0;
}
--
2.17.2
^ permalink raw reply related
* [PATCH V2 net-next 04/11] net: ena: enable the interrupt_moderation in driver_supported_features
From: akiyano @ 2019-09-16 11:31 UTC (permalink / raw)
To: davem, netdev
Cc: Arthur Kiyanovski, dwmw, zorik, matua, saeedb, msw, aliguori,
nafea, gtzalik, netanel, alisaidi, benh, sameehj, ndagan
In-Reply-To: <1568633496-4143-1-git-send-email-akiyano@amazon.com>
From: Arthur Kiyanovski <akiyano@amazon.com>
Add driver_supported_features to host_host info which is a new API used to
communicate to the device which features are supported by the driver.
Add the interrupt_moderation bit to host_info->driver_supported_features
and enable it to signal the device that this driver supports interrupt
moderation properly.
Reserved bits are for features implemented in the future
Signed-off-by: Arthur Kiyanovski <akiyano@amazon.com>
---
drivers/net/ethernet/amazon/ena/ena_admin_defs.h | 8 ++++++++
drivers/net/ethernet/amazon/ena/ena_netdev.c | 3 +++
2 files changed, 11 insertions(+)
diff --git a/drivers/net/ethernet/amazon/ena/ena_admin_defs.h b/drivers/net/ethernet/amazon/ena/ena_admin_defs.h
index d19f2ecf8e84..8baf847e8622 100644
--- a/drivers/net/ethernet/amazon/ena/ena_admin_defs.h
+++ b/drivers/net/ethernet/amazon/ena/ena_admin_defs.h
@@ -808,6 +808,12 @@ struct ena_admin_host_info {
u16 num_cpus;
u16 reserved;
+
+ /* 1 :0 : reserved
+ * 2 : interrupt_moderation
+ * 31:3 : reserved
+ */
+ u32 driver_supported_features;
};
struct ena_admin_rss_ind_table_entry {
@@ -1110,6 +1116,8 @@ struct ena_admin_ena_mmio_req_read_less_resp {
#define ENA_ADMIN_HOST_INFO_DEVICE_MASK GENMASK(7, 3)
#define ENA_ADMIN_HOST_INFO_BUS_SHIFT 8
#define ENA_ADMIN_HOST_INFO_BUS_MASK GENMASK(15, 8)
+#define ENA_ADMIN_HOST_INFO_INTERRUPT_MODERATION_SHIFT 2
+#define ENA_ADMIN_HOST_INFO_INTERRUPT_MODERATION_MASK BIT(2)
/* aenq_common_desc */
#define ENA_ADMIN_AENQ_COMMON_DESC_PHASE_MASK BIT(0)
diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.c b/drivers/net/ethernet/amazon/ena/ena_netdev.c
index cdcc169b87fa..f19736493c01 100644
--- a/drivers/net/ethernet/amazon/ena/ena_netdev.c
+++ b/drivers/net/ethernet/amazon/ena/ena_netdev.c
@@ -2438,6 +2438,9 @@ static void ena_config_host_info(struct ena_com_dev *ena_dev,
("K"[0] << ENA_ADMIN_HOST_INFO_MODULE_TYPE_SHIFT);
host_info->num_cpus = num_online_cpus();
+ host_info->driver_supported_features =
+ ENA_ADMIN_HOST_INFO_INTERRUPT_MODERATION_MASK;
+
rc = ena_com_set_host_attributes(ena_dev);
if (rc) {
if (rc == -EOPNOTSUPP)
--
2.17.2
^ permalink raw reply related
* [PATCH V2 net-next 05/11] net: ena: remove code duplication in ena_com_update_nonadaptive_moderation_interval _*()
From: akiyano @ 2019-09-16 11:31 UTC (permalink / raw)
To: davem, netdev
Cc: Arthur Kiyanovski, dwmw, zorik, matua, saeedb, msw, aliguori,
nafea, gtzalik, netanel, alisaidi, benh, sameehj, ndagan
In-Reply-To: <1568633496-4143-1-git-send-email-akiyano@amazon.com>
From: Arthur Kiyanovski <akiyano@amazon.com>
Remove code duplication in:
ena_com_update_nonadaptive_moderation_interval_tx()
ena_com_update_nonadaptive_moderation_interval_rx()
functions.
Signed-off-by: Arthur Kiyanovski <akiyano@amazon.com>
---
drivers/net/ethernet/amazon/ena/ena_com.c | 30 ++++++++++++-----------
1 file changed, 16 insertions(+), 14 deletions(-)
diff --git a/drivers/net/ethernet/amazon/ena/ena_com.c b/drivers/net/ethernet/amazon/ena/ena_com.c
index f3c67c27c19c..bbb59fd220a1 100644
--- a/drivers/net/ethernet/amazon/ena/ena_com.c
+++ b/drivers/net/ethernet/amazon/ena/ena_com.c
@@ -2772,32 +2772,34 @@ bool ena_com_interrupt_moderation_supported(struct ena_com_dev *ena_dev)
ENA_ADMIN_INTERRUPT_MODERATION);
}
-int ena_com_update_nonadaptive_moderation_interval_tx(struct ena_com_dev *ena_dev,
- u32 tx_coalesce_usecs)
+static int ena_com_update_nonadaptive_moderation_interval(u32 coalesce_usecs,
+ u32 intr_delay_resolution,
+ u32 *intr_moder_interval)
{
- if (!ena_dev->intr_delay_resolution) {
+ if (!intr_delay_resolution) {
pr_err("Illegal interrupt delay granularity value\n");
return -EFAULT;
}
- ena_dev->intr_moder_tx_interval = tx_coalesce_usecs /
- ena_dev->intr_delay_resolution;
+ *intr_moder_interval = coalesce_usecs / intr_delay_resolution;
return 0;
}
+int ena_com_update_nonadaptive_moderation_interval_tx(struct ena_com_dev *ena_dev,
+ u32 tx_coalesce_usecs)
+{
+ return ena_com_update_nonadaptive_moderation_interval(tx_coalesce_usecs,
+ ena_dev->intr_delay_resolution,
+ &ena_dev->intr_moder_tx_interval);
+}
+
int ena_com_update_nonadaptive_moderation_interval_rx(struct ena_com_dev *ena_dev,
u32 rx_coalesce_usecs)
{
- if (!ena_dev->intr_delay_resolution) {
- pr_err("Illegal interrupt delay granularity value\n");
- return -EFAULT;
- }
-
- ena_dev->intr_moder_rx_interval = rx_coalesce_usecs /
- ena_dev->intr_delay_resolution;
-
- return 0;
+ return ena_com_update_nonadaptive_moderation_interval(rx_coalesce_usecs,
+ ena_dev->intr_delay_resolution,
+ &ena_dev->intr_moder_rx_interval);
}
void ena_com_destroy_interrupt_moderation(struct ena_com_dev *ena_dev)
--
2.17.2
^ permalink raw reply related
* [PATCH V2 net-next 06/11] net: ena: remove old adaptive interrupt moderation code from ena_netdev
From: akiyano @ 2019-09-16 11:31 UTC (permalink / raw)
To: davem, netdev
Cc: Arthur Kiyanovski, dwmw, zorik, matua, saeedb, msw, aliguori,
nafea, gtzalik, netanel, alisaidi, benh, sameehj, ndagan
In-Reply-To: <1568633496-4143-1-git-send-email-akiyano@amazon.com>
From: Arthur Kiyanovski <akiyano@amazon.com>
1. Out of the fields {per_napi_bytes, per_napi_packets} in struct ena_ring,
only rx_ring->per_napi_packets are used to determine if napi did work
for dim.
This commit removes all other uses of these fields.
2. Remove ena_ring->moder_tbl_idx, which is not used by dim.
3. Remove all calls to ena_com_destroy_interrupt_moderation(), since all it
did was to destroy the interrupt moderation table, which is removed as
part of removing old interrupt moderation code.
Signed-off-by: Arthur Kiyanovski <akiyano@amazon.com>
---
drivers/net/ethernet/amazon/ena/ena_netdev.c | 8 --------
drivers/net/ethernet/amazon/ena/ena_netdev.h | 2 --
2 files changed, 10 deletions(-)
diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.c b/drivers/net/ethernet/amazon/ena/ena_netdev.c
index f19736493c01..1d8855ab8624 100644
--- a/drivers/net/ethernet/amazon/ena/ena_netdev.c
+++ b/drivers/net/ethernet/amazon/ena/ena_netdev.c
@@ -158,7 +158,6 @@ static void ena_init_io_rings_common(struct ena_adapter *adapter,
ring->adapter = adapter;
ring->ena_dev = adapter->ena_dev;
ring->per_napi_packets = 0;
- ring->per_napi_bytes = 0;
ring->cpu = 0;
ring->first_interrupt = false;
ring->no_interrupt_event_cnt = 0;
@@ -834,9 +833,6 @@ static int ena_clean_tx_irq(struct ena_ring *tx_ring, u32 budget)
__netif_tx_unlock(txq);
}
- tx_ring->per_napi_bytes += tx_bytes;
- tx_ring->per_napi_packets += tx_pkts;
-
return tx_pkts;
}
@@ -1120,7 +1116,6 @@ static int ena_clean_rx_irq(struct ena_ring *rx_ring, struct napi_struct *napi,
} while (likely(res_budget));
work_done = budget - res_budget;
- rx_ring->per_napi_bytes += total_len;
rx_ring->per_napi_packets += work_done;
u64_stats_update_begin(&rx_ring->syncp);
rx_ring->rx_stats.bytes += total_len;
@@ -3641,7 +3636,6 @@ static int ena_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
ena_free_mgmnt_irq(adapter);
ena_disable_msix(adapter);
err_worker_destroy:
- ena_com_destroy_interrupt_moderation(ena_dev);
del_timer(&adapter->timer_service);
err_netdev_destroy:
free_netdev(netdev);
@@ -3702,8 +3696,6 @@ static void ena_remove(struct pci_dev *pdev)
pci_disable_device(pdev);
- ena_com_destroy_interrupt_moderation(ena_dev);
-
vfree(ena_dev);
}
diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.h b/drivers/net/ethernet/amazon/ena/ena_netdev.h
index f67ecab3389a..e9450991ab74 100644
--- a/drivers/net/ethernet/amazon/ena/ena_netdev.h
+++ b/drivers/net/ethernet/amazon/ena/ena_netdev.h
@@ -280,8 +280,6 @@ struct ena_ring {
struct ena_com_rx_buf_info ena_bufs[ENA_PKT_MAX_BUFS];
u32 smoothed_interval;
u32 per_napi_packets;
- u32 per_napi_bytes;
- enum ena_intr_moder_level moder_tbl_idx;
u16 non_empty_napi_events;
struct u64_stats_sync syncp;
union {
--
2.17.2
^ permalink raw reply related
* [PATCH V2 net-next 07/11] net: ena: remove ena_restore_ethtool_params() and relevant fields
From: akiyano @ 2019-09-16 11:31 UTC (permalink / raw)
To: davem, netdev
Cc: Arthur Kiyanovski, dwmw, zorik, matua, saeedb, msw, aliguori,
nafea, gtzalik, netanel, alisaidi, benh, sameehj, ndagan
In-Reply-To: <1568633496-4143-1-git-send-email-akiyano@amazon.com>
From: Arthur Kiyanovski <akiyano@amazon.com>
Deleted unused 4 fields from struct ena_adapter and their only user
ena_restore_ethtool_params().
Signed-off-by: Arthur Kiyanovski <akiyano@amazon.com>
---
drivers/net/ethernet/amazon/ena/ena_netdev.c | 10 ----------
drivers/net/ethernet/amazon/ena/ena_netdev.h | 3 ---
2 files changed, 13 deletions(-)
diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.c b/drivers/net/ethernet/amazon/ena/ena_netdev.c
index 1d8855ab8624..8b9f8b90e525 100644
--- a/drivers/net/ethernet/amazon/ena/ena_netdev.c
+++ b/drivers/net/ethernet/amazon/ena/ena_netdev.c
@@ -1563,14 +1563,6 @@ static void ena_napi_enable_all(struct ena_adapter *adapter)
napi_enable(&adapter->ena_napi[i].napi);
}
-static void ena_restore_ethtool_params(struct ena_adapter *adapter)
-{
- adapter->tx_usecs = 0;
- adapter->rx_usecs = 0;
- adapter->tx_frames = 1;
- adapter->rx_frames = 1;
-}
-
/* Configure the Rx forwarding */
static int ena_rss_configure(struct ena_adapter *adapter)
{
@@ -1620,8 +1612,6 @@ static int ena_up_complete(struct ena_adapter *adapter)
/* enable transmits */
netif_tx_start_all_queues(adapter->netdev);
- ena_restore_ethtool_params(adapter);
-
ena_napi_enable_all(adapter);
return 0;
diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.h b/drivers/net/ethernet/amazon/ena/ena_netdev.h
index e9450991ab74..72ee51a82ec7 100644
--- a/drivers/net/ethernet/amazon/ena/ena_netdev.h
+++ b/drivers/net/ethernet/amazon/ena/ena_netdev.h
@@ -330,9 +330,6 @@ struct ena_adapter {
u32 missing_tx_completion_threshold;
- u32 tx_usecs, rx_usecs; /* interrupt moderation */
- u32 tx_frames, rx_frames; /* interrupt moderation */
-
u32 requested_tx_ring_size;
u32 requested_rx_ring_size;
--
2.17.2
^ permalink raw reply related
* [PATCH V2 net-next 09/11] net: ena: fix update of interrupt moderation register
From: akiyano @ 2019-09-16 11:31 UTC (permalink / raw)
To: davem, netdev
Cc: Arthur Kiyanovski, dwmw, zorik, matua, saeedb, msw, aliguori,
nafea, gtzalik, netanel, alisaidi, benh, sameehj, ndagan
In-Reply-To: <1568633496-4143-1-git-send-email-akiyano@amazon.com>
From: Arthur Kiyanovski <akiyano@amazon.com>
Current implementation always updates the interrupt register with
the smoothed_interval of the rx_ring. However this should be
done only in case of adaptive interrupt moderation. If non-adaptive
interrupt moderation is used, the non-adaptive interrupt moderation
interval should be used. This commit fixes that.
Signed-off-by: Arthur Kiyanovski <akiyano@amazon.com>
---
drivers/net/ethernet/amazon/ena/ena_netdev.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.c b/drivers/net/ethernet/amazon/ena/ena_netdev.c
index 8b9f8b90e525..54539e57aa73 100644
--- a/drivers/net/ethernet/amazon/ena/ena_netdev.c
+++ b/drivers/net/ethernet/amazon/ena/ena_netdev.c
@@ -1187,12 +1187,15 @@ static void ena_unmask_interrupt(struct ena_ring *tx_ring,
struct ena_ring *rx_ring)
{
struct ena_eth_io_intr_reg intr_reg;
+ u32 rx_interval = ena_com_get_adaptive_moderation_enabled(rx_ring->ena_dev) ?
+ rx_ring->smoothed_interval :
+ ena_com_get_nonadaptive_moderation_interval_rx(rx_ring->ena_dev);
/* Update intr register: rx intr delay,
* tx intr delay and interrupt unmask
*/
ena_com_update_intr_reg(&intr_reg,
- rx_ring->smoothed_interval,
+ rx_interval,
tx_ring->smoothed_interval,
true);
--
2.17.2
^ permalink raw reply related
* [PATCH V2 net-next 08/11] net: ena: remove all old adaptive rx interrupt moderation code from ena_com
From: akiyano @ 2019-09-16 11:31 UTC (permalink / raw)
To: davem, netdev
Cc: Arthur Kiyanovski, dwmw, zorik, matua, saeedb, msw, aliguori,
nafea, gtzalik, netanel, alisaidi, benh, sameehj, ndagan
In-Reply-To: <1568633496-4143-1-git-send-email-akiyano@amazon.com>
From: Arthur Kiyanovski <akiyano@amazon.com>
Remove previous implementation of adaptive rx interrupt moderation
from ena_com files.
Signed-off-by: Arthur Kiyanovski <akiyano@amazon.com>
---
drivers/net/ethernet/amazon/ena/ena_com.c | 110 -----------------
drivers/net/ethernet/amazon/ena/ena_com.h | 142 ----------------------
2 files changed, 252 deletions(-)
diff --git a/drivers/net/ethernet/amazon/ena/ena_com.c b/drivers/net/ethernet/amazon/ena/ena_com.c
index bbb59fd220a1..621b747f062b 100644
--- a/drivers/net/ethernet/amazon/ena/ena_com.c
+++ b/drivers/net/ethernet/amazon/ena/ena_com.c
@@ -1278,22 +1278,6 @@ static int ena_com_ind_tbl_convert_from_device(struct ena_com_dev *ena_dev)
return 0;
}
-static int ena_com_init_interrupt_moderation_table(struct ena_com_dev *ena_dev)
-{
- size_t size;
-
- size = sizeof(struct ena_intr_moder_entry) * ENA_INTR_MAX_NUM_OF_LEVELS;
-
- ena_dev->intr_moder_tbl =
- devm_kzalloc(ena_dev->dmadev, size, GFP_KERNEL);
- if (!ena_dev->intr_moder_tbl)
- return -ENOMEM;
-
- ena_com_config_default_interrupt_moderation_table(ena_dev);
-
- return 0;
-}
-
static void ena_com_update_intr_delay_resolution(struct ena_com_dev *ena_dev,
u16 intr_delay_resolution)
{
@@ -2802,13 +2786,6 @@ int ena_com_update_nonadaptive_moderation_interval_rx(struct ena_com_dev *ena_de
&ena_dev->intr_moder_rx_interval);
}
-void ena_com_destroy_interrupt_moderation(struct ena_com_dev *ena_dev)
-{
- if (ena_dev->intr_moder_tbl)
- devm_kfree(ena_dev->dmadev, ena_dev->intr_moder_tbl);
- ena_dev->intr_moder_tbl = NULL;
-}
-
int ena_com_init_interrupt_moderation(struct ena_com_dev *ena_dev)
{
struct ena_admin_get_feat_resp get_resp;
@@ -2833,10 +2810,6 @@ int ena_com_init_interrupt_moderation(struct ena_com_dev *ena_dev)
return rc;
}
- rc = ena_com_init_interrupt_moderation_table(ena_dev);
- if (rc)
- goto err;
-
/* if moderation is supported by device we set adaptive moderation */
delay_resolution = get_resp.u.intr_moderation.intr_delay_resolution;
ena_com_update_intr_delay_resolution(ena_dev, delay_resolution);
@@ -2845,52 +2818,6 @@ int ena_com_init_interrupt_moderation(struct ena_com_dev *ena_dev)
ena_com_disable_adaptive_moderation(ena_dev);
return 0;
-err:
- ena_com_destroy_interrupt_moderation(ena_dev);
- return rc;
-}
-
-void ena_com_config_default_interrupt_moderation_table(struct ena_com_dev *ena_dev)
-{
- struct ena_intr_moder_entry *intr_moder_tbl = ena_dev->intr_moder_tbl;
-
- if (!intr_moder_tbl)
- return;
-
- intr_moder_tbl[ENA_INTR_MODER_LOWEST].intr_moder_interval =
- ENA_INTR_LOWEST_USECS;
- intr_moder_tbl[ENA_INTR_MODER_LOWEST].pkts_per_interval =
- ENA_INTR_LOWEST_PKTS;
- intr_moder_tbl[ENA_INTR_MODER_LOWEST].bytes_per_interval =
- ENA_INTR_LOWEST_BYTES;
-
- intr_moder_tbl[ENA_INTR_MODER_LOW].intr_moder_interval =
- ENA_INTR_LOW_USECS;
- intr_moder_tbl[ENA_INTR_MODER_LOW].pkts_per_interval =
- ENA_INTR_LOW_PKTS;
- intr_moder_tbl[ENA_INTR_MODER_LOW].bytes_per_interval =
- ENA_INTR_LOW_BYTES;
-
- intr_moder_tbl[ENA_INTR_MODER_MID].intr_moder_interval =
- ENA_INTR_MID_USECS;
- intr_moder_tbl[ENA_INTR_MODER_MID].pkts_per_interval =
- ENA_INTR_MID_PKTS;
- intr_moder_tbl[ENA_INTR_MODER_MID].bytes_per_interval =
- ENA_INTR_MID_BYTES;
-
- intr_moder_tbl[ENA_INTR_MODER_HIGH].intr_moder_interval =
- ENA_INTR_HIGH_USECS;
- intr_moder_tbl[ENA_INTR_MODER_HIGH].pkts_per_interval =
- ENA_INTR_HIGH_PKTS;
- intr_moder_tbl[ENA_INTR_MODER_HIGH].bytes_per_interval =
- ENA_INTR_HIGH_BYTES;
-
- intr_moder_tbl[ENA_INTR_MODER_HIGHEST].intr_moder_interval =
- ENA_INTR_HIGHEST_USECS;
- intr_moder_tbl[ENA_INTR_MODER_HIGHEST].pkts_per_interval =
- ENA_INTR_HIGHEST_PKTS;
- intr_moder_tbl[ENA_INTR_MODER_HIGHEST].bytes_per_interval =
- ENA_INTR_HIGHEST_BYTES;
}
unsigned int ena_com_get_nonadaptive_moderation_interval_tx(struct ena_com_dev *ena_dev)
@@ -2903,43 +2830,6 @@ unsigned int ena_com_get_nonadaptive_moderation_interval_rx(struct ena_com_dev *
return ena_dev->intr_moder_rx_interval;
}
-void ena_com_init_intr_moderation_entry(struct ena_com_dev *ena_dev,
- enum ena_intr_moder_level level,
- struct ena_intr_moder_entry *entry)
-{
- struct ena_intr_moder_entry *intr_moder_tbl = ena_dev->intr_moder_tbl;
-
- if (level >= ENA_INTR_MAX_NUM_OF_LEVELS)
- return;
-
- intr_moder_tbl[level].intr_moder_interval = entry->intr_moder_interval;
- if (ena_dev->intr_delay_resolution)
- intr_moder_tbl[level].intr_moder_interval /=
- ena_dev->intr_delay_resolution;
- intr_moder_tbl[level].pkts_per_interval = entry->pkts_per_interval;
-
- /* use hardcoded value until ethtool supports bytecount parameter */
- if (entry->bytes_per_interval != ENA_INTR_BYTE_COUNT_NOT_SUPPORTED)
- intr_moder_tbl[level].bytes_per_interval = entry->bytes_per_interval;
-}
-
-void ena_com_get_intr_moderation_entry(struct ena_com_dev *ena_dev,
- enum ena_intr_moder_level level,
- struct ena_intr_moder_entry *entry)
-{
- struct ena_intr_moder_entry *intr_moder_tbl = ena_dev->intr_moder_tbl;
-
- if (level >= ENA_INTR_MAX_NUM_OF_LEVELS)
- return;
-
- entry->intr_moder_interval = intr_moder_tbl[level].intr_moder_interval;
- if (ena_dev->intr_delay_resolution)
- entry->intr_moder_interval *= ena_dev->intr_delay_resolution;
- entry->pkts_per_interval =
- intr_moder_tbl[level].pkts_per_interval;
- entry->bytes_per_interval = intr_moder_tbl[level].bytes_per_interval;
-}
-
int ena_com_config_dev_mode(struct ena_com_dev *ena_dev,
struct ena_admin_feature_llq_desc *llq_features,
struct ena_llq_configurations *llq_default_cfg)
diff --git a/drivers/net/ethernet/amazon/ena/ena_com.h b/drivers/net/ethernet/amazon/ena/ena_com.h
index baeefc6af4f3..ddc2a8c50333 100644
--- a/drivers/net/ethernet/amazon/ena/ena_com.h
+++ b/drivers/net/ethernet/amazon/ena/ena_com.h
@@ -72,46 +72,13 @@
/*****************************************************************************/
/* ENA adaptive interrupt moderation settings */
-#define ENA_INTR_LOWEST_USECS (0)
-#define ENA_INTR_LOWEST_PKTS (3)
-#define ENA_INTR_LOWEST_BYTES (2 * 1524)
-
-#define ENA_INTR_LOW_USECS (32)
-#define ENA_INTR_LOW_PKTS (12)
-#define ENA_INTR_LOW_BYTES (16 * 1024)
-
-#define ENA_INTR_MID_USECS (80)
-#define ENA_INTR_MID_PKTS (48)
-#define ENA_INTR_MID_BYTES (64 * 1024)
-
-#define ENA_INTR_HIGH_USECS (128)
-#define ENA_INTR_HIGH_PKTS (96)
-#define ENA_INTR_HIGH_BYTES (128 * 1024)
-
-#define ENA_INTR_HIGHEST_USECS (192)
-#define ENA_INTR_HIGHEST_PKTS (128)
-#define ENA_INTR_HIGHEST_BYTES (192 * 1024)
-
#define ENA_INTR_INITIAL_TX_INTERVAL_USECS 196
#define ENA_INTR_INITIAL_RX_INTERVAL_USECS 0
-#define ENA_INTR_DELAY_OLD_VALUE_WEIGHT 6
-#define ENA_INTR_DELAY_NEW_VALUE_WEIGHT 4
-#define ENA_INTR_MODER_LEVEL_STRIDE 2
-#define ENA_INTR_BYTE_COUNT_NOT_SUPPORTED 0xFFFFFF
#define ENA_HW_HINTS_NO_TIMEOUT 0xFFFF
#define ENA_FEATURE_MAX_QUEUE_EXT_VER 1
-enum ena_intr_moder_level {
- ENA_INTR_MODER_LOWEST = 0,
- ENA_INTR_MODER_LOW,
- ENA_INTR_MODER_MID,
- ENA_INTR_MODER_HIGH,
- ENA_INTR_MODER_HIGHEST,
- ENA_INTR_MAX_NUM_OF_LEVELS,
-};
-
struct ena_llq_configurations {
enum ena_admin_llq_header_location llq_header_location;
enum ena_admin_llq_ring_entry_size llq_ring_entry_size;
@@ -120,12 +87,6 @@ struct ena_llq_configurations {
u16 llq_ring_entry_size_value;
};
-struct ena_intr_moder_entry {
- unsigned int intr_moder_interval;
- unsigned int pkts_per_interval;
- unsigned int bytes_per_interval;
-};
-
enum queue_direction {
ENA_COM_IO_QUEUE_DIRECTION_TX,
ENA_COM_IO_QUEUE_DIRECTION_RX
@@ -920,11 +881,6 @@ int ena_com_execute_admin_command(struct ena_com_admin_queue *admin_queue,
*/
int ena_com_init_interrupt_moderation(struct ena_com_dev *ena_dev);
-/* ena_com_destroy_interrupt_moderation - Destroy interrupt moderation resources
- * @ena_dev: ENA communication layer struct
- */
-void ena_com_destroy_interrupt_moderation(struct ena_com_dev *ena_dev);
-
/* ena_com_interrupt_moderation_supported - Return if interrupt moderation
* capability is supported by the device.
*
@@ -932,12 +888,6 @@ void ena_com_destroy_interrupt_moderation(struct ena_com_dev *ena_dev);
*/
bool ena_com_interrupt_moderation_supported(struct ena_com_dev *ena_dev);
-/* ena_com_config_default_interrupt_moderation_table - Restore the interrupt
- * moderation table back to the default parameters.
- * @ena_dev: ENA communication layer struct
- */
-void ena_com_config_default_interrupt_moderation_table(struct ena_com_dev *ena_dev);
-
/* ena_com_update_nonadaptive_moderation_interval_tx - Update the
* non-adaptive interval in Tx direction.
* @ena_dev: ENA communication layer struct
@@ -974,29 +924,6 @@ unsigned int ena_com_get_nonadaptive_moderation_interval_tx(struct ena_com_dev *
*/
unsigned int ena_com_get_nonadaptive_moderation_interval_rx(struct ena_com_dev *ena_dev);
-/* ena_com_init_intr_moderation_entry - Update a single entry in the interrupt
- * moderation table.
- * @ena_dev: ENA communication layer struct
- * @level: Interrupt moderation table level
- * @entry: Entry value
- *
- * Update a single entry in the interrupt moderation table.
- */
-void ena_com_init_intr_moderation_entry(struct ena_com_dev *ena_dev,
- enum ena_intr_moder_level level,
- struct ena_intr_moder_entry *entry);
-
-/* ena_com_get_intr_moderation_entry - Init ena_intr_moder_entry.
- * @ena_dev: ENA communication layer struct
- * @level: Interrupt moderation table level
- * @entry: Entry to fill.
- *
- * Initialize the entry according to the adaptive interrupt moderation table.
- */
-void ena_com_get_intr_moderation_entry(struct ena_com_dev *ena_dev,
- enum ena_intr_moder_level level,
- struct ena_intr_moder_entry *entry);
-
/* ena_com_config_dev_mode - Configure the placement policy of the device.
* @ena_dev: ENA communication layer struct
* @llq_features: LLQ feature descriptor, retrieve via
@@ -1022,75 +949,6 @@ static inline void ena_com_disable_adaptive_moderation(struct ena_com_dev *ena_d
ena_dev->adaptive_coalescing = false;
}
-/* ena_com_calculate_interrupt_delay - Calculate new interrupt delay
- * @ena_dev: ENA communication layer struct
- * @pkts: Number of packets since the last update
- * @bytes: Number of bytes received since the last update.
- * @smoothed_interval: Returned interval
- * @moder_tbl_idx: Current table level as input update new level as return
- * value.
- */
-static inline void ena_com_calculate_interrupt_delay(struct ena_com_dev *ena_dev,
- unsigned int pkts,
- unsigned int bytes,
- unsigned int *smoothed_interval,
- unsigned int *moder_tbl_idx)
-{
- enum ena_intr_moder_level curr_moder_idx, new_moder_idx;
- struct ena_intr_moder_entry *curr_moder_entry;
- struct ena_intr_moder_entry *pred_moder_entry;
- struct ena_intr_moder_entry *new_moder_entry;
- struct ena_intr_moder_entry *intr_moder_tbl = ena_dev->intr_moder_tbl;
- unsigned int interval;
-
- /* We apply adaptive moderation on Rx path only.
- * Tx uses static interrupt moderation.
- */
- if (!pkts || !bytes)
- /* Tx interrupt, or spurious interrupt,
- * in both cases we just use same delay values
- */
- return;
-
- curr_moder_idx = (enum ena_intr_moder_level)(*moder_tbl_idx);
- if (unlikely(curr_moder_idx >= ENA_INTR_MAX_NUM_OF_LEVELS)) {
- pr_err("Wrong moderation index %u\n", curr_moder_idx);
- return;
- }
-
- curr_moder_entry = &intr_moder_tbl[curr_moder_idx];
- new_moder_idx = curr_moder_idx;
-
- if (curr_moder_idx == ENA_INTR_MODER_LOWEST) {
- if ((pkts > curr_moder_entry->pkts_per_interval) ||
- (bytes > curr_moder_entry->bytes_per_interval))
- new_moder_idx =
- (enum ena_intr_moder_level)(curr_moder_idx + ENA_INTR_MODER_LEVEL_STRIDE);
- } else {
- pred_moder_entry = &intr_moder_tbl[curr_moder_idx - ENA_INTR_MODER_LEVEL_STRIDE];
-
- if ((pkts <= pred_moder_entry->pkts_per_interval) ||
- (bytes <= pred_moder_entry->bytes_per_interval))
- new_moder_idx =
- (enum ena_intr_moder_level)(curr_moder_idx - ENA_INTR_MODER_LEVEL_STRIDE);
- else if ((pkts > curr_moder_entry->pkts_per_interval) ||
- (bytes > curr_moder_entry->bytes_per_interval)) {
- if (curr_moder_idx != ENA_INTR_MODER_HIGHEST)
- new_moder_idx =
- (enum ena_intr_moder_level)(curr_moder_idx + ENA_INTR_MODER_LEVEL_STRIDE);
- }
- }
- new_moder_entry = &intr_moder_tbl[new_moder_idx];
-
- interval = new_moder_entry->intr_moder_interval;
- *smoothed_interval = (
- (interval * ENA_INTR_DELAY_NEW_VALUE_WEIGHT +
- ENA_INTR_DELAY_OLD_VALUE_WEIGHT * (*smoothed_interval)) + 5) /
- 10;
-
- *moder_tbl_idx = new_moder_idx;
-}
-
/* ena_com_update_intr_reg - Prepare interrupt register
* @intr_reg: interrupt register to update.
* @rx_delay_interval: Rx interval in usecs
--
2.17.2
^ permalink raw reply related
* [PATCH V2 net-next 02/11] net: ena: switch to dim algorithm for rx adaptive interrupt moderation
From: akiyano @ 2019-09-16 11:31 UTC (permalink / raw)
To: davem, netdev
Cc: Arthur Kiyanovski, dwmw, zorik, matua, saeedb, msw, aliguori,
nafea, gtzalik, netanel, alisaidi, benh, sameehj, ndagan
In-Reply-To: <1568633496-4143-1-git-send-email-akiyano@amazon.com>
From: Arthur Kiyanovski <akiyano@amazon.com>
Use the dim library for the rx adaptive interrupt moderation implementation
Signed-off-by: Arthur Kiyanovski <akiyano@amazon.com>
---
drivers/net/ethernet/amazon/ena/ena_com.c | 4 +-
drivers/net/ethernet/amazon/ena/ena_netdev.c | 55 +++++++++++++-------
drivers/net/ethernet/amazon/ena/ena_netdev.h | 3 ++
3 files changed, 41 insertions(+), 21 deletions(-)
diff --git a/drivers/net/ethernet/amazon/ena/ena_com.c b/drivers/net/ethernet/amazon/ena/ena_com.c
index 4144936bf8a2..f3c67c27c19c 100644
--- a/drivers/net/ethernet/amazon/ena/ena_com.c
+++ b/drivers/net/ethernet/amazon/ena/ena_com.c
@@ -2839,9 +2839,7 @@ int ena_com_init_interrupt_moderation(struct ena_com_dev *ena_dev)
delay_resolution = get_resp.u.intr_moderation.intr_delay_resolution;
ena_com_update_intr_delay_resolution(ena_dev, delay_resolution);
- /* Disable adaptive moderation by default - can be enabled from
- * ethtool
- */
+ /* Disable adaptive moderation by default - can be enabled later */
ena_com_disable_adaptive_moderation(ena_dev);
return 0;
diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.c b/drivers/net/ethernet/amazon/ena/ena_netdev.c
index 233b252ceb9e..cdcc169b87fa 100644
--- a/drivers/net/ethernet/amazon/ena/ena_netdev.c
+++ b/drivers/net/ethernet/amazon/ena/ena_netdev.c
@@ -196,6 +196,7 @@ static void ena_init_io_rings(struct ena_adapter *adapter)
rxr->smoothed_interval =
ena_com_get_nonadaptive_moderation_interval_rx(ena_dev);
rxr->empty_rx_queue = 0;
+ adapter->ena_napi[i].dim.mode = DIM_CQ_PERIOD_MODE_START_FROM_EQE;
}
}
@@ -712,6 +713,7 @@ static void ena_destroy_all_rx_queues(struct ena_adapter *adapter)
for (i = 0; i < adapter->num_queues; i++) {
ena_qid = ENA_IO_RXQ_IDX(i);
+ cancel_work_sync(&adapter->ena_napi[i].dim.work);
ena_com_destroy_io_queue(adapter->ena_dev, ena_qid);
}
}
@@ -1155,23 +1157,35 @@ static int ena_clean_rx_irq(struct ena_ring *rx_ring, struct napi_struct *napi,
return 0;
}
-void ena_adjust_intr_moderation(struct ena_ring *rx_ring,
- struct ena_ring *tx_ring)
+static void ena_dim_work(struct work_struct *w)
{
- /* We apply adaptive moderation on Rx path only.
- * Tx uses static interrupt moderation.
- */
- ena_com_calculate_interrupt_delay(rx_ring->ena_dev,
- rx_ring->per_napi_packets,
- rx_ring->per_napi_bytes,
- &rx_ring->smoothed_interval,
- &rx_ring->moder_tbl_idx);
-
- /* Reset per napi packets/bytes */
- tx_ring->per_napi_packets = 0;
- tx_ring->per_napi_bytes = 0;
+ struct dim *dim = container_of(w, struct dim, work);
+ struct dim_cq_moder cur_moder =
+ net_dim_get_rx_moderation(dim->mode, dim->profile_ix);
+ struct ena_napi *ena_napi = container_of(dim, struct ena_napi, dim);
+
+ ena_napi->rx_ring->smoothed_interval = cur_moder.usec;
+ dim->state = DIM_START_MEASURE;
+}
+
+static void ena_adjust_adaptive_rx_intr_moderation(struct ena_napi *ena_napi)
+{
+ struct dim_sample dim_sample;
+ struct ena_ring *rx_ring = ena_napi->rx_ring;
+
+ if (!rx_ring->per_napi_packets)
+ return;
+
+ rx_ring->non_empty_napi_events++;
+
+ dim_update_sample(rx_ring->non_empty_napi_events,
+ rx_ring->rx_stats.cnt,
+ rx_ring->rx_stats.bytes,
+ &dim_sample);
+
+ net_dim(&ena_napi->dim, dim_sample);
+
rx_ring->per_napi_packets = 0;
- rx_ring->per_napi_bytes = 0;
}
static void ena_unmask_interrupt(struct ena_ring *tx_ring,
@@ -1260,9 +1274,11 @@ static int ena_io_poll(struct napi_struct *napi, int budget)
* from the interrupt context (vs from sk_busy_loop)
*/
if (napi_complete_done(napi, rx_work_done)) {
- /* Tx and Rx share the same interrupt vector */
+ /* We apply adaptive moderation on Rx path only.
+ * Tx uses static interrupt moderation.
+ */
if (ena_com_get_adaptive_moderation_enabled(rx_ring->ena_dev))
- ena_adjust_intr_moderation(rx_ring, tx_ring);
+ ena_adjust_adaptive_rx_intr_moderation(ena_napi);
ena_unmask_interrupt(tx_ring, rx_ring);
}
@@ -1740,13 +1756,16 @@ static int ena_create_all_io_rx_queues(struct ena_adapter *adapter)
rc = ena_create_io_rx_queue(adapter, i);
if (rc)
goto create_err;
+ INIT_WORK(&adapter->ena_napi[i].dim.work, ena_dim_work);
}
return 0;
create_err:
- while (i--)
+ while (i--) {
+ cancel_work_sync(&adapter->ena_napi[i].dim.work);
ena_com_destroy_io_queue(ena_dev, ENA_IO_RXQ_IDX(i));
+ }
return rc;
}
diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.h b/drivers/net/ethernet/amazon/ena/ena_netdev.h
index efbcffd22215..f67ecab3389a 100644
--- a/drivers/net/ethernet/amazon/ena/ena_netdev.h
+++ b/drivers/net/ethernet/amazon/ena/ena_netdev.h
@@ -34,6 +34,7 @@
#define ENA_H
#include <linux/bitops.h>
+#include <linux/dim.h>
#include <linux/etherdevice.h>
#include <linux/inetdevice.h>
#include <linux/interrupt.h>
@@ -153,6 +154,7 @@ struct ena_napi {
struct ena_ring *tx_ring;
struct ena_ring *rx_ring;
u32 qid;
+ struct dim dim;
};
struct ena_calc_queue_size_ctx {
@@ -280,6 +282,7 @@ struct ena_ring {
u32 per_napi_packets;
u32 per_napi_bytes;
enum ena_intr_moder_level moder_tbl_idx;
+ u16 non_empty_napi_events;
struct u64_stats_sync syncp;
union {
struct ena_stats_tx tx_stats;
--
2.17.2
^ permalink raw reply related
* [PATCH V2 net-next 11/11] net: ena: fix incorrect update of intr_delay_resolution
From: akiyano @ 2019-09-16 11:31 UTC (permalink / raw)
To: davem, netdev
Cc: Arthur Kiyanovski, dwmw, zorik, matua, saeedb, msw, aliguori,
nafea, gtzalik, netanel, alisaidi, benh, sameehj, ndagan
In-Reply-To: <1568633496-4143-1-git-send-email-akiyano@amazon.com>
From: Arthur Kiyanovski <akiyano@amazon.com>
ena_dev->intr_moder_rx/tx_interval save the intervals received from the
user after dividing them by ena_dev->intr_delay_resolution. Therefore
when intr_delay_resolution changes, the code needs to first mutiply
intr_moder_rx/tx_interval by the previous intr_delay_resolution to get
the value originally given by the user, and only then divide it by the
new intr_delay_resolution.
Current code does not first multiply intr_moder_rx/tx_interval by the old
intr_delay_resolution. This commit fixes it.
Also initialize ena_dev->intr_delay_resolution to be 1.
Signed-off-by: Arthur Kiyanovski <akiyano@amazon.com>
---
drivers/net/ethernet/amazon/ena/ena_com.c | 21 ++++++++++++++++----
drivers/net/ethernet/amazon/ena/ena_com.h | 1 +
drivers/net/ethernet/amazon/ena/ena_netdev.c | 1 +
3 files changed, 19 insertions(+), 4 deletions(-)
diff --git a/drivers/net/ethernet/amazon/ena/ena_com.c b/drivers/net/ethernet/amazon/ena/ena_com.c
index 621b747f062b..ea62604fdf8c 100644
--- a/drivers/net/ethernet/amazon/ena/ena_com.c
+++ b/drivers/net/ethernet/amazon/ena/ena_com.c
@@ -1281,17 +1281,30 @@ static int ena_com_ind_tbl_convert_from_device(struct ena_com_dev *ena_dev)
static void ena_com_update_intr_delay_resolution(struct ena_com_dev *ena_dev,
u16 intr_delay_resolution)
{
+ /* Initial value of intr_delay_resolution might be 0 */
+ u16 prev_intr_delay_resolution =
+ ena_dev->intr_delay_resolution ?
+ ena_dev->intr_delay_resolution :
+ ENA_DEFAULT_INTR_DELAY_RESOLUTION;
+
if (!intr_delay_resolution) {
pr_err("Illegal intr_delay_resolution provided. Going to use default 1 usec resolution\n");
- intr_delay_resolution = 1;
+ intr_delay_resolution = ENA_DEFAULT_INTR_DELAY_RESOLUTION;
}
- ena_dev->intr_delay_resolution = intr_delay_resolution;
/* update Rx */
- ena_dev->intr_moder_rx_interval /= intr_delay_resolution;
+ ena_dev->intr_moder_rx_interval =
+ ena_dev->intr_moder_rx_interval *
+ prev_intr_delay_resolution /
+ intr_delay_resolution;
/* update Tx */
- ena_dev->intr_moder_tx_interval /= intr_delay_resolution;
+ ena_dev->intr_moder_tx_interval =
+ ena_dev->intr_moder_tx_interval *
+ prev_intr_delay_resolution /
+ intr_delay_resolution;
+
+ ena_dev->intr_delay_resolution = intr_delay_resolution;
}
/*****************************************************************************/
diff --git a/drivers/net/ethernet/amazon/ena/ena_com.h b/drivers/net/ethernet/amazon/ena/ena_com.h
index ddc2a8c50333..7c941eba0bc9 100644
--- a/drivers/net/ethernet/amazon/ena/ena_com.h
+++ b/drivers/net/ethernet/amazon/ena/ena_com.h
@@ -74,6 +74,7 @@
#define ENA_INTR_INITIAL_TX_INTERVAL_USECS 196
#define ENA_INTR_INITIAL_RX_INTERVAL_USECS 0
+#define ENA_DEFAULT_INTR_DELAY_RESOLUTION 1
#define ENA_HW_HINTS_NO_TIMEOUT 0xFFFF
diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.c b/drivers/net/ethernet/amazon/ena/ena_netdev.c
index 54539e57aa73..e4bf7a4af87a 100644
--- a/drivers/net/ethernet/amazon/ena/ena_netdev.c
+++ b/drivers/net/ethernet/amazon/ena/ena_netdev.c
@@ -3500,6 +3500,7 @@ static int ena_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
*/
ena_dev->intr_moder_tx_interval = ENA_INTR_INITIAL_TX_INTERVAL_USECS;
ena_dev->intr_moder_rx_interval = ENA_INTR_INITIAL_RX_INTERVAL_USECS;
+ ena_dev->intr_delay_resolution = ENA_DEFAULT_INTR_DELAY_RESOLUTION;
io_queue_num = ena_calc_io_queue_num(pdev, ena_dev, &get_feat_ctx);
rc = ena_calc_queue_size(&calc_queue_ctx);
if (rc || io_queue_num <= 0) {
--
2.17.2
^ permalink raw reply related
* [PATCH V2 net-next 00/11] net: ena: implement adaptive interrupt moderation using dim
From: akiyano @ 2019-09-16 11:31 UTC (permalink / raw)
To: davem, netdev
Cc: Arthur Kiyanovski, dwmw, zorik, matua, saeedb, msw, aliguori,
nafea, gtzalik, netanel, alisaidi, benh, sameehj, ndagan
From: Arthur Kiyanovski <akiyano@amazon.com>
In this patchset we replace our adaptive interrupt moderation
implementation with the dim library implementation.
The dim library showed great improvement in throughput, latency
and CPU usage in different scenarios on ARM CPUs.
This patchset also includes a few bug fixes to the parts of the
old implementation of adaptive interrupt moderation that were left.
Changes from V1 patchset:
Removed stray empty lines from patches 01/11, 09/11.
Arthur Kiyanovski (11):
net: ena: add intr_moder_rx_interval to struct ena_com_dev and use it
net: ena: switch to dim algorithm for rx adaptive interrupt moderation
net: ena: reimplement set/get_coalesce()
net: ena: enable the interrupt_moderation in driver_supported_features
net: ena: remove code duplication in
ena_com_update_nonadaptive_moderation_interval _*()
net: ena: remove old adaptive interrupt moderation code from
ena_netdev
net: ena: remove ena_restore_ethtool_params() and relevant fields
net: ena: remove all old adaptive rx interrupt moderation code from
ena_com
net: ena: fix update of interrupt moderation register
net: ena: fix retrieval of nonadaptive interrupt moderation intervals
net: ena: fix incorrect update of intr_delay_resolution
.../net/ethernet/amazon/ena/ena_admin_defs.h | 8 +
drivers/net/ethernet/amazon/ena/ena_com.c | 175 ++++--------------
drivers/net/ethernet/amazon/ena/ena_com.h | 151 +--------------
drivers/net/ethernet/amazon/ena/ena_ethtool.c | 89 +++------
drivers/net/ethernet/amazon/ena/ena_netdev.c | 85 +++++----
drivers/net/ethernet/amazon/ena/ena_netdev.h | 8 +-
6 files changed, 128 insertions(+), 388 deletions(-)
--
2.17.2
^ permalink raw reply
* [PATCH V2 net-next 10/11] net: ena: fix retrieval of nonadaptive interrupt moderation intervals
From: akiyano @ 2019-09-16 11:31 UTC (permalink / raw)
To: davem, netdev
Cc: Arthur Kiyanovski, dwmw, zorik, matua, saeedb, msw, aliguori,
nafea, gtzalik, netanel, alisaidi, benh, sameehj, ndagan
In-Reply-To: <1568633496-4143-1-git-send-email-akiyano@amazon.com>
From: Arthur Kiyanovski <akiyano@amazon.com>
Nonadaptive interrupt moderation intervals are assigned the value set
by the user in ethtool -C divided by ena_dev->intr_delay_resolution.
Therefore when the user tries to get the nonadaptive interrupt moderation
intervals with ethtool -c the code needs to multiply the saved value
by ena_dev->intr_delay_resolution.
The current code erroneously divides instead of multiplying in ethtool -c.
This patch fixes this.
Signed-off-by: Arthur Kiyanovski <akiyano@amazon.com>
---
drivers/net/ethernet/amazon/ena/ena_ethtool.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ethernet/amazon/ena/ena_ethtool.c b/drivers/net/ethernet/amazon/ena/ena_ethtool.c
index 0f90e2296630..16553d92fad2 100644
--- a/drivers/net/ethernet/amazon/ena/ena_ethtool.c
+++ b/drivers/net/ethernet/amazon/ena/ena_ethtool.c
@@ -310,14 +310,15 @@ static int ena_get_coalesce(struct net_device *net_dev,
/* the devie doesn't support interrupt moderation */
return -EOPNOTSUPP;
}
+
coalesce->tx_coalesce_usecs =
- ena_com_get_nonadaptive_moderation_interval_tx(ena_dev) /
+ ena_com_get_nonadaptive_moderation_interval_tx(ena_dev) *
ena_dev->intr_delay_resolution;
if (!ena_com_get_adaptive_moderation_enabled(ena_dev))
coalesce->rx_coalesce_usecs =
ena_com_get_nonadaptive_moderation_interval_rx(ena_dev)
- / ena_dev->intr_delay_resolution;
+ * ena_dev->intr_delay_resolution;
coalesce->use_adaptive_rx_coalesce =
ena_com_get_adaptive_moderation_enabled(ena_dev);
--
2.17.2
^ permalink raw reply related
* RE: [PATCH V1 net-next 01/11] net: ena: add intr_moder_rx_interval to struct ena_com_dev and use it
From: Kiyanovski, Arthur @ 2019-09-16 11:39 UTC (permalink / raw)
To: David Miller
Cc: netdev@vger.kernel.org, Woodhouse, David, Machulsky, Zorik,
Matushevsky, Alexander, Bshara, Saeed, Wilson, Matt,
Liguori, Anthony, Bshara, Nafea, Tzalik, Guy, Belgazal, Netanel,
Saidi, Ali, Herrenschmidt, Benjamin, Jubran, Samih, Dagan, Noam
In-Reply-To: <20190915.193241.878202512573492759.davem@davemloft.net>
> -----Original Message-----
> From: David Miller <davem@davemloft.net>
> Sent: Sunday, September 15, 2019 9:33 PM
> To: Kiyanovski, Arthur <akiyano@amazon.com>
> Cc: netdev@vger.kernel.org; Woodhouse, David <dwmw@amazon.co.uk>;
> Machulsky, Zorik <zorik@amazon.com>; Matushevsky, Alexander
> <matua@amazon.com>; Bshara, Saeed <saeedb@amazon.com>; Wilson, Matt
> <msw@amazon.com>; Liguori, Anthony <aliguori@amazon.com>; Bshara,
> Nafea <nafea@amazon.com>; Tzalik, Guy <gtzalik@amazon.com>; Belgazal,
> Netanel <netanel@amazon.com>; Saidi, Ali <alisaidi@amazon.com>;
> Herrenschmidt, Benjamin <benh@amazon.com>; Jubran, Samih
> <sameehj@amazon.com>; Dagan, Noam <ndagan@amazon.com>
> Subject: Re: [PATCH V1 net-next 01/11] net: ena: add intr_moder_rx_interval to
> struct ena_com_dev and use it
>
> From: <akiyano@amazon.com>
> Date: Fri, 13 Sep 2019 01:08:38 +0300
>
> > @@ -1307,8 +1304,8 @@ static void
> ena_com_update_intr_delay_resolution(struct ena_com_dev *ena_dev,
> > ena_dev->intr_delay_resolution = intr_delay_resolution;
> >
> > /* update Rx */
> > - for (i = 0; i < ENA_INTR_MAX_NUM_OF_LEVELS; i++)
> > - intr_moder_tbl[i].intr_moder_interval /= intr_delay_resolution;
> > + ena_dev->intr_moder_rx_interval /= intr_delay_resolution;
> > +
> >
> > /* update Tx */
>
> Now there are two empty lines here, please remove one of them.
Thanks.
For some reason checkpatch did not catch this and also another such extra space.
Sent out V2 of the patchset.
^ permalink raw reply
* Re: [PATCH v5 2/2] net: phy: adin: implement Energy Detect Powerdown mode via phy-tunable
From: Andrew Lunn @ 2019-09-16 11:57 UTC (permalink / raw)
To: Alexandru Ardelean
Cc: netdev, devicetree, linux-kernel, davem, robh+dt, mark.rutland,
f.fainelli, hkallweit1, mkubecek
In-Reply-To: <20190916073526.24711-3-alexandru.ardelean@analog.com>
On Mon, Sep 16, 2019 at 10:35:26AM +0300, Alexandru Ardelean wrote:
> This driver becomes the first user of the kernel's `ETHTOOL_PHY_EDPD`
> phy-tunable feature.
> EDPD is also enabled by default on PHY config_init, but can be disabled via
> the phy-tunable control.
>
> When enabling EDPD, it's also a good idea (for the ADIN PHYs) to enable TX
> periodic pulses, so that in case the other PHY is also on EDPD mode, there
> is no lock-up situation where both sides are waiting for the other to
> transmit.
>
> Via the phy-tunable control, TX pulses can be disabled if specifying 0
> `tx-interval` via ethtool.
>
> The ADIN PHY supports only fixed 1 second intervals; they cannot be
> configured. That is why the acceptable values are 1,
> ETHTOOL_PHY_EDPD_DFLT_TX_MSECS and ETHTOOL_PHY_EDPD_NO_TX (which disables
> TX pulses).
>
> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
> Signed-off-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Andrew
^ permalink raw reply
* [PATCH] selftests: Add test cases for `ip nexthop flush proto XX`
From: Donald Sharp @ 2019-09-16 12:26 UTC (permalink / raw)
To: netdev, dsahern
Add some test cases to allow the fib_nexthops.sh test code
to test the flushing of nexthops based upon the proto passed
in upon creation of the nexthop group.
Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
---
tools/testing/selftests/net/fib_nexthops.sh | 14 ++++++++++++++
1 file changed, 14 insertions(+)
diff --git a/tools/testing/selftests/net/fib_nexthops.sh b/tools/testing/selftests/net/fib_nexthops.sh
index f9ebeac1e6f2..796670ebc65b 100755
--- a/tools/testing/selftests/net/fib_nexthops.sh
+++ b/tools/testing/selftests/net/fib_nexthops.sh
@@ -940,6 +940,20 @@ basic()
run_cmd "$IP nexthop add id 104 group 1 dev veth1"
log_test $? 2 "Nexthop group and device"
+ # Tests to ensure that flushing works as expected.
+ run_cmd "$IP nexthop add id 105 blackhole proto 99"
+ run_cmd "$IP nexthop add id 106 blackhole proto 100"
+ run_cmd "$IP nexthop add id 107 blackhole proto 99"
+ run_cmd "$IP nexthop flush proto 99"
+ check_nexthop "id 105" ""
+ check_nexthop "id 106" "id 106 blackhole proto 100"
+ check_nexthop "id 107" ""
+ run_cmd "$IP nexthop flush proto 100"
+ check_nexthop "id 106" ""
+
+ run_cmd "$IP nexthop flush proto 100"
+ log_test $? 0 "Test proto flush"
+
run_cmd "$IP nexthop add id 104 group 1 blackhole"
log_test $? 2 "Nexthop group and blackhole"
--
2.21.0
^ permalink raw reply related
* [PATCH iproute2-next v2] bpf: replace snprintf with asprintf when dealing with long buffers
From: Andrea Claudi @ 2019-09-16 13:00 UTC (permalink / raw)
To: netdev; +Cc: stephen, dsahern
This reduces stack usage, as asprintf allocates memory on the heap.
This indirectly fixes a snprintf truncation warning (from gcc v9.2.1):
bpf.c: In function ‘bpf_get_work_dir’:
bpf.c:784:49: warning: ‘snprintf’ output may be truncated before the last format character [-Wformat-truncation=]
784 | snprintf(bpf_wrk_dir, sizeof(bpf_wrk_dir), "%s/", mnt);
| ^
bpf.c:784:2: note: ‘snprintf’ output between 2 and 4097 bytes into a destination of size 4096
784 | snprintf(bpf_wrk_dir, sizeof(bpf_wrk_dir), "%s/", mnt);
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Fixes: e42256699cac ("bpf: make tc's bpf loader generic and move into lib")
Signed-off-by: Andrea Claudi <aclaudi@redhat.com>
---
lib/bpf.c | 155 ++++++++++++++++++++++++++++++++++++++++--------------
1 file changed, 116 insertions(+), 39 deletions(-)
diff --git a/lib/bpf.c b/lib/bpf.c
index 7d2a322ffbaec..0fe2e9eb2cbf8 100644
--- a/lib/bpf.c
+++ b/lib/bpf.c
@@ -406,13 +406,21 @@ static int bpf_derive_elf_map_from_fdinfo(int fd, struct bpf_elf_map *map,
struct bpf_map_ext *ext)
{
unsigned int val, owner_type = 0, owner_jited = 0;
- char file[PATH_MAX], buff[4096];
+ char *file = NULL;
+ char buff[4096];
FILE *fp;
+ int ret;
- snprintf(file, sizeof(file), "/proc/%d/fdinfo/%d", getpid(), fd);
+ ret = asprintf(&file, "/proc/%d/fdinfo/%d", getpid(), fd);
+ if (ret < 0) {
+ fprintf(stderr, "asprintf failed: %s\n", strerror(errno));
+ free(file);
+ return ret;
+ }
memset(map, 0, sizeof(*map));
fp = fopen(file, "r");
+ free(file);
if (!fp) {
fprintf(stderr, "No procfs support?!\n");
return -EIO;
@@ -600,8 +608,9 @@ int bpf_trace_pipe(void)
0,
};
int fd_in, fd_out = STDERR_FILENO;
- char tpipe[PATH_MAX];
+ char *tpipe = NULL;
const char *mnt;
+ int ret;
mnt = bpf_find_mntpt("tracefs", TRACEFS_MAGIC, tracefs_mnt,
sizeof(tracefs_mnt), tracefs_known_mnts);
@@ -610,9 +619,15 @@ int bpf_trace_pipe(void)
return -1;
}
- snprintf(tpipe, sizeof(tpipe), "%s/trace_pipe", mnt);
+ ret = asprintf(&tpipe, "%s/trace_pipe", mnt);
+ if (ret < 0) {
+ fprintf(stderr, "asprintf failed: %s\n", strerror(errno));
+ free(tpipe);
+ return ret;
+ }
fd_in = open(tpipe, O_RDONLY);
+ free(tpipe);
if (fd_in < 0)
return -1;
@@ -633,37 +648,50 @@ int bpf_trace_pipe(void)
static int bpf_gen_global(const char *bpf_sub_dir)
{
- char bpf_glo_dir[PATH_MAX];
+ char *bpf_glo_dir = NULL;
int ret;
- snprintf(bpf_glo_dir, sizeof(bpf_glo_dir), "%s/%s/",
- bpf_sub_dir, BPF_DIR_GLOBALS);
+ ret = asprintf(&bpf_glo_dir, "%s/%s/", bpf_sub_dir, BPF_DIR_GLOBALS);
+ if (ret < 0) {
+ fprintf(stderr, "asprintf failed: %s\n", strerror(errno));
+ goto out;
+ }
ret = mkdir(bpf_glo_dir, S_IRWXU);
if (ret && errno != EEXIST) {
fprintf(stderr, "mkdir %s failed: %s\n", bpf_glo_dir,
strerror(errno));
- return ret;
+ goto out;
}
- return 0;
+ ret = 0;
+out:
+ free(bpf_glo_dir);
+ return ret;
}
static int bpf_gen_master(const char *base, const char *name)
{
- char bpf_sub_dir[PATH_MAX + NAME_MAX + 1];
+ char *bpf_sub_dir = NULL;
int ret;
- snprintf(bpf_sub_dir, sizeof(bpf_sub_dir), "%s%s/", base, name);
+ ret = asprintf(&bpf_sub_dir, "%s%s/", base, name);
+ if (ret < 0) {
+ fprintf(stderr, "asprintf failed: %s\n", strerror(errno));
+ goto out;
+ }
ret = mkdir(bpf_sub_dir, S_IRWXU);
if (ret && errno != EEXIST) {
fprintf(stderr, "mkdir %s failed: %s\n", bpf_sub_dir,
strerror(errno));
- return ret;
+ goto out;
}
- return bpf_gen_global(bpf_sub_dir);
+ ret = bpf_gen_global(bpf_sub_dir);
+out:
+ free(bpf_sub_dir);
+ return ret;
}
static int bpf_slave_via_bind_mnt(const char *full_name,
@@ -692,13 +720,22 @@ static int bpf_slave_via_bind_mnt(const char *full_name,
static int bpf_gen_slave(const char *base, const char *name,
const char *link)
{
- char bpf_lnk_dir[PATH_MAX + NAME_MAX + 1];
- char bpf_sub_dir[PATH_MAX + NAME_MAX];
+ char *bpf_lnk_dir = NULL;
+ char *bpf_sub_dir = NULL;
struct stat sb = {};
int ret;
- snprintf(bpf_lnk_dir, sizeof(bpf_lnk_dir), "%s%s/", base, link);
- snprintf(bpf_sub_dir, sizeof(bpf_sub_dir), "%s%s", base, name);
+ ret = asprintf(&bpf_lnk_dir, "%s%s/", base, link);
+ if (ret < 0) {
+ fprintf(stderr, "asprintf failed: %s\n", strerror(errno));
+ goto out;
+ }
+
+ ret = asprintf(&bpf_sub_dir, "%s%s", base, name);
+ if (ret < 0) {
+ fprintf(stderr, "asprintf failed: %s\n", strerror(errno));
+ goto out;
+ }
ret = symlink(bpf_lnk_dir, bpf_sub_dir);
if (ret) {
@@ -706,25 +743,30 @@ static int bpf_gen_slave(const char *base, const char *name,
if (errno != EPERM) {
fprintf(stderr, "symlink %s failed: %s\n",
bpf_sub_dir, strerror(errno));
- return ret;
+ goto out;
}
- return bpf_slave_via_bind_mnt(bpf_sub_dir,
- bpf_lnk_dir);
+ ret = bpf_slave_via_bind_mnt(bpf_sub_dir, bpf_lnk_dir);
+ goto out;
}
ret = lstat(bpf_sub_dir, &sb);
if (ret) {
fprintf(stderr, "lstat %s failed: %s\n",
bpf_sub_dir, strerror(errno));
- return ret;
+ goto out;
}
- if ((sb.st_mode & S_IFMT) != S_IFLNK)
- return bpf_gen_global(bpf_sub_dir);
+ if ((sb.st_mode & S_IFMT) != S_IFLNK) {
+ ret = bpf_gen_global(bpf_sub_dir);
+ goto out;
+ }
}
- return 0;
+out:
+ free(bpf_lnk_dir);
+ free(bpf_sub_dir);
+ return ret;
}
static int bpf_gen_hierarchy(const char *base)
@@ -742,7 +784,7 @@ static int bpf_gen_hierarchy(const char *base)
static const char *bpf_get_work_dir(enum bpf_prog_type type)
{
static char bpf_tmp[PATH_MAX] = BPF_DIR_MNT;
- static char bpf_wrk_dir[PATH_MAX];
+ static char *bpf_wrk_dir;
static const char *mnt;
static bool bpf_mnt_cached;
const char *mnt_env = getenv(BPF_ENV_MNT);
@@ -781,7 +823,12 @@ static const char *bpf_get_work_dir(enum bpf_prog_type type)
}
}
- snprintf(bpf_wrk_dir, sizeof(bpf_wrk_dir), "%s/", mnt);
+ ret = asprintf(&bpf_wrk_dir, "%s/", mnt);
+ if (ret < 0) {
+ fprintf(stderr, "asprintf failed: %s\n", strerror(errno));
+ free(bpf_wrk_dir);
+ goto out;
+ }
ret = bpf_gen_hierarchy(bpf_wrk_dir);
if (ret) {
@@ -1438,31 +1485,48 @@ static int bpf_probe_pinned(const char *name, const struct bpf_elf_ctx *ctx,
static int bpf_make_obj_path(const struct bpf_elf_ctx *ctx)
{
- char tmp[PATH_MAX];
+ char *tmp = NULL;
int ret;
- snprintf(tmp, sizeof(tmp), "%s/%s", bpf_get_work_dir(ctx->type),
- ctx->obj_uid);
+ ret = asprintf(&tmp, "%s/%s", bpf_get_work_dir(ctx->type), ctx->obj_uid);
+ if (ret < 0) {
+ fprintf(stderr, "asprintf failed: %s\n", strerror(errno));
+ goto out;
+ }
ret = mkdir(tmp, S_IRWXU);
if (ret && errno != EEXIST) {
fprintf(stderr, "mkdir %s failed: %s\n", tmp, strerror(errno));
- return ret;
+ goto out;
}
- return 0;
+ ret = 0;
+out:
+ free(tmp);
+ return ret;
}
static int bpf_make_custom_path(const struct bpf_elf_ctx *ctx,
const char *todo)
{
- char tmp[PATH_MAX], rem[PATH_MAX], *sub;
+ char *tmp = NULL;
+ char *rem = NULL;
+ char *sub;
int ret;
- snprintf(tmp, sizeof(tmp), "%s/../", bpf_get_work_dir(ctx->type));
- snprintf(rem, sizeof(rem), "%s/", todo);
- sub = strtok(rem, "/");
+ ret = asprintf(&tmp, "%s/../", bpf_get_work_dir(ctx->type));
+ if (ret < 0) {
+ fprintf(stderr, "asprintf failed: %s\n", strerror(errno));
+ goto out;
+ }
+ ret = asprintf(&rem, "%s/", todo);
+ if (ret < 0) {
+ fprintf(stderr, "asprintf failed: %s\n", strerror(errno));
+ goto out;
+ }
+
+ sub = strtok(rem, "/");
while (sub) {
if (strlen(tmp) + strlen(sub) + 2 > PATH_MAX)
return -EINVAL;
@@ -1474,13 +1538,17 @@ static int bpf_make_custom_path(const struct bpf_elf_ctx *ctx,
if (ret && errno != EEXIST) {
fprintf(stderr, "mkdir %s failed: %s\n", tmp,
strerror(errno));
- return ret;
+ goto out;
}
sub = strtok(NULL, "/");
}
- return 0;
+ ret = 0;
+out:
+ free(rem);
+ free(tmp);
+ return ret;
}
static int bpf_place_pinned(int fd, const char *name,
@@ -2581,14 +2649,23 @@ struct bpf_jited_aux {
static int bpf_derive_prog_from_fdinfo(int fd, struct bpf_prog_data *prog)
{
- char file[PATH_MAX], buff[4096];
+ char *file = NULL;
+ char buff[4096];
unsigned int val;
FILE *fp;
+ int ret;
+
+ ret = asprintf(&file, "/proc/%d/fdinfo/%d", getpid(), fd);
+ if (ret < 0) {
+ fprintf(stderr, "asprintf failed: %s\n", strerror(errno));
+ free(file);
+ return ret;
+ }
- snprintf(file, sizeof(file), "/proc/%d/fdinfo/%d", getpid(), fd);
memset(prog, 0, sizeof(*prog));
fp = fopen(file, "r");
+ free(file);
if (!fp) {
fprintf(stderr, "No procfs support?!\n");
return -EIO;
--
2.21.0
^ permalink raw reply related
* [PATCH bpf v2] libbpf: Remove getsockopt() check for XDP_OPTIONS
From: Toke Høiland-Jørgensen @ 2019-09-16 12:33 UTC (permalink / raw)
To: Alexei Starovoitov, Daniel Borkmann, netdev, bpf
Cc: Björn Töpel, Toke Høiland-Jørgensen,
Yonghong Song
The xsk_socket__create() function fails and returns an error if it cannot
get the XDP_OPTIONS through getsockopt(). However, support for XDP_OPTIONS
was not added until kernel 5.3, so this means that creating XSK sockets
always fails on older kernels.
Since the option is just used to set the zero-copy flag in the xsk struct,
and that flag is not really used for anything yet, just remove the
getsockopt() call until a proper use for it is introduced.
Suggested-by: Yonghong Song <yhs@fb.com>
Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
---
v2:
- Remove the call entirely.
tools/lib/bpf/xsk.c | 11 -----------
1 file changed, 11 deletions(-)
diff --git a/tools/lib/bpf/xsk.c b/tools/lib/bpf/xsk.c
index 680e63066cf3..14aea315c914 100644
--- a/tools/lib/bpf/xsk.c
+++ b/tools/lib/bpf/xsk.c
@@ -65,7 +65,6 @@ struct xsk_socket {
int xsks_map_fd;
__u32 queue_id;
char ifname[IFNAMSIZ];
- bool zc;
};
struct xsk_nl_info {
@@ -481,7 +480,6 @@ int xsk_socket__create(struct xsk_socket **xsk_ptr, const char *ifname,
void *rx_map = NULL, *tx_map = NULL;
struct sockaddr_xdp sxdp = {};
struct xdp_mmap_offsets off;
- struct xdp_options opts;
struct xsk_socket *xsk;
socklen_t optlen;
int err;
@@ -601,15 +599,6 @@ int xsk_socket__create(struct xsk_socket **xsk_ptr, const char *ifname,
xsk->prog_fd = -1;
- optlen = sizeof(opts);
- err = getsockopt(xsk->fd, SOL_XDP, XDP_OPTIONS, &opts, &optlen);
- if (err) {
- err = -errno;
- goto out_mmap_tx;
- }
-
- xsk->zc = opts.flags & XDP_OPTIONS_ZEROCOPY;
-
if (!(xsk->config.libbpf_flags & XSK_LIBBPF_FLAGS__INHIBIT_PROG_LOAD)) {
err = xsk_setup_xdp_prog(xsk);
if (err)
--
2.23.0
^ permalink raw reply related
* Re: [v3 2/3] samples: pktgen: add helper functions for IP(v4/v6) CIDR parsing
From: Toke Høiland-Jørgensen @ 2019-09-16 12:52 UTC (permalink / raw)
To: Jesper Dangaard Brouer, Daniel T. Lee; +Cc: David S . Miller, netdev, brouer
In-Reply-To: <20190916085317.02e4d985@carbon>
Jesper Dangaard Brouer <brouer@redhat.com> writes:
> On Sun, 15 Sep 2019 00:13:52 +0900
> "Daniel T. Lee" <danieltimlee@gmail.com> wrote:
>
>> This commit adds CIDR parsing and IP validate helper function to parse
>> single IP or range of IP with CIDR. (e.g. 198.18.0.0/15)
>>
>> Helpers will be used in prior to set target address in samples/pktgen.
>>
>> Signed-off-by: Daniel T. Lee <danieltimlee@gmail.com>
>> ---
>> Changes since v3:
>> * Set errexit option to stop script execution on error
>>
>> samples/pktgen/functions.sh | 124 ++++++++++++++++++++++++++++++++++++
>> 1 file changed, 124 insertions(+)
>>
>> diff --git a/samples/pktgen/functions.sh b/samples/pktgen/functions.sh
>> index 4af4046d71be..87ae61701904 100644
>> --- a/samples/pktgen/functions.sh
>> +++ b/samples/pktgen/functions.sh
>> @@ -5,6 +5,8 @@
>> # Author: Jesper Dangaaard Brouer
>> # License: GPL
>>
>> +set -o errexit
>
> Unfortunately, this breaks the scripts.
>
> The function proc_cmd are designed to grep after "Result: OK:" which
> might fail, and your patch/change makes the script stop immediately.
> We actually want to continue, and output what command that failed (and
> also grep again after "Result:" to provide the kernel reason).
>
> Even if you somehow "fix" function proc_cmd, then we in general want to
> catch different error situations by looking at status $?, and output
> meaning full errors via calling err() function.
Yeah, I can see that some functions do this, but I don't think that
would be too hard to fix. See sample diff below (will need some more
work to deal with grep failing, but you get the idea).
I'd argue that fixing this is the right thing to do... Maybe add set -o
nounset as well while we're at it? :)
> IHMO as minimum with errexit you need a 'trap' function that can
> help/inform the user of what went wrong.
Yeah, trap ERR (which would also need set -o errtrace to work inside the
functions) might be useful in any case.
-Toke
diff --git a/samples/pktgen/functions.sh b/samples/pktgen/functions.sh
index 4af4046d71be..d61a348f85f5 100644
--- a/samples/pktgen/functions.sh
+++ b/samples/pktgen/functions.sh
@@ -58,6 +58,7 @@ function pg_set() {
function proc_cmd() {
local result
local proc_file=$1
+ local status=0
# after shift, the remaining args are contained in $@
shift
local proc_ctrl=${PROC_DIR}/$proc_file
@@ -73,8 +74,7 @@ function proc_cmd() {
echo "cmd: $@ > $proc_ctrl"
fi
# Quoting of "$@" is important for space expansion
- echo "$@" > "$proc_ctrl"
- local status=$?
+ echo "$@" > "$proc_ctrl" || status=$?
result=$(grep "Result: OK:" $proc_ctrl)
# Due to pgctrl, cannot use exit code $? from grep
^ permalink raw reply related
* Re: [PATCH] netfilter: bridge: drop a broken include
From: Jeremy Sowden @ 2019-09-16 13:08 UTC (permalink / raw)
To: Adam Borowski
Cc: Pablo Neira Ayuso, Jozsef Kadlecsik, Florian Westphal,
Roopa Prabhu, Nikolay Aleksandrov, netfilter-devel, coreteam,
netdev
In-Reply-To: <20190916000517.45028-1-kilobyte@angband.pl>
[-- Attachment #1: Type: text/plain, Size: 4755 bytes --]
On 2019-09-16, at 02:05:16 +0200, Adam Borowski wrote:
> This caused a build failure if CONFIG_NF_CONNTRACK_BRIDGE is set but
> CONFIG_NF_TABLES=n -- and appears to be unused anyway.
>
> Signed-off-by: Adam Borowski <kilobyte@angband.pl>
> ---
> net/bridge/netfilter/nf_conntrack_bridge.c | 1 -
> 1 file changed, 1 deletion(-)
>
> diff --git a/net/bridge/netfilter/nf_conntrack_bridge.c
> b/net/bridge/netfilter/nf_conntrack_bridge.c
> index 4f5444d2a526..844ef5a53f87 100644
> --- a/net/bridge/netfilter/nf_conntrack_bridge.c
> +++ b/net/bridge/netfilter/nf_conntrack_bridge.c
> @@ -18,7 +18,6 @@
>
> #include <linux/netfilter/nf_tables.h>
> #include <net/netfilter/ipv6/nf_defrag_ipv6.h>
> -#include <net/netfilter/nf_tables.h>
>
> #include "../br_private.h"
>
> --
> 2.23.0
This is the build-failure (reproduced with v5.3):
In file included from net/bridge/netfilter/nf_conntrack_bridge.c:21:
./include/net/netfilter/nf_tables.h: In function ‘nft_gencursor_next’:
./include/net/netfilter/nf_tables.h:1223:14: error: ‘const struct net’ has no member named ‘nft’; did you mean ‘nf’?
1223 | return net->nft.gencursor + 1 == 1 ? 1 : 0;
| ^~~
| nf
In file included from ./include/linux/export.h:45,
from ./include/linux/linkage.h:7,
from ./include/linux/kernel.h:8,
from ./include/linux/skbuff.h:13,
from ./include/linux/ip.h:16,
from net/bridge/netfilter/nf_conntrack_bridge.c:3:
./include/net/netfilter/nf_tables.h: In function ‘nft_genmask_cur’:
./include/net/netfilter/nf_tables.h:1234:29: error: ‘const struct net’ has no member named ‘nft’; did you mean ‘nf’?
1234 | return 1 << READ_ONCE(net->nft.gencursor);
| ^~~
./include/linux/compiler.h:261:17: note: in definition of macro ‘__READ_ONCE’
261 | union { typeof(x) __val; char __c[1]; } __u; \
| ^
./include/net/netfilter/nf_tables.h:1234:14: note: in expansion of macro ‘READ_ONCE’
1234 | return 1 << READ_ONCE(net->nft.gencursor);
| ^~~~~~~~~
./include/net/netfilter/nf_tables.h:1234:29: error: ‘const struct net’ has no member named ‘nft’; did you mean ‘nf’?
1234 | return 1 << READ_ONCE(net->nft.gencursor);
| ^~~
./include/linux/compiler.h:263:22: note: in definition of macro ‘__READ_ONCE’
263 | __read_once_size(&(x), __u.__c, sizeof(x)); \
| ^
./include/net/netfilter/nf_tables.h:1234:14: note: in expansion of macro ‘READ_ONCE’
1234 | return 1 << READ_ONCE(net->nft.gencursor);
| ^~~~~~~~~
./include/net/netfilter/nf_tables.h:1234:29: error: ‘const struct net’ has no member named ‘nft’; did you mean ‘nf’?
1234 | return 1 << READ_ONCE(net->nft.gencursor);
| ^~~
./include/linux/compiler.h:263:42: note: in definition of macro ‘__READ_ONCE’
263 | __read_once_size(&(x), __u.__c, sizeof(x)); \
| ^
./include/net/netfilter/nf_tables.h:1234:14: note: in expansion of macro ‘READ_ONCE’
1234 | return 1 << READ_ONCE(net->nft.gencursor);
| ^~~~~~~~~
./include/net/netfilter/nf_tables.h:1234:29: error: ‘const struct net’ has no member named ‘nft’; did you mean ‘nf’?
1234 | return 1 << READ_ONCE(net->nft.gencursor);
| ^~~
./include/linux/compiler.h:265:30: note: in definition of macro ‘__READ_ONCE’
265 | __read_once_size_nocheck(&(x), __u.__c, sizeof(x)); \
| ^
./include/net/netfilter/nf_tables.h:1234:14: note: in expansion of macro ‘READ_ONCE’
1234 | return 1 << READ_ONCE(net->nft.gencursor);
| ^~~~~~~~~
./include/net/netfilter/nf_tables.h:1234:29: error: ‘const struct net’ has no member named ‘nft’; did you mean ‘nf’?
1234 | return 1 << READ_ONCE(net->nft.gencursor);
| ^~~
./include/linux/compiler.h:265:50: note: in definition of macro ‘__READ_ONCE’
265 | __read_once_size_nocheck(&(x), __u.__c, sizeof(x)); \
| ^
./include/net/netfilter/nf_tables.h:1234:14: note: in expansion of macro ‘READ_ONCE’
1234 | return 1 << READ_ONCE(net->nft.gencursor);
| ^~~~~~~~~
There are already changes in the net-next tree that will fix it.
J.
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]
^ permalink raw reply
* Re: [PATCH] libbpf: Don't error out if getsockopt() fails for XDP_OPTIONS
From: Toke Høiland-Jørgensen @ 2019-09-16 12:31 UTC (permalink / raw)
To: Daniel Borkmann, Yonghong Song, Alexei Starovoitov,
netdev@vger.kernel.org, bpf@vger.kernel.org, maximmi@mellanox.com
In-Reply-To: <9271a44f-1bbf-1305-bff9-8cbb8bae9098@iogearbox.net>
Daniel Borkmann <daniel@iogearbox.net> writes:
> On 9/13/19 8:53 PM, Yonghong Song wrote:
>> On 9/10/19 12:06 AM, Toke Høiland-Jørgensen wrote:
>>> Yonghong Song <yhs@fb.com> writes:
>>>> On 9/9/19 10:46 AM, Toke Høiland-Jørgensen wrote:
>>>>> The xsk_socket__create() function fails and returns an error if it cannot
>>>>> get the XDP_OPTIONS through getsockopt(). However, support for XDP_OPTIONS
>>>>> was not added until kernel 5.3, so this means that creating XSK sockets
>>>>> always fails on older kernels.
>>>>>
>>>>> Since the option is just used to set the zero-copy flag in the xsk struct,
>>>>> there really is no need to error out if the getsockopt() call fails.
>>>>>
>>>>> Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
>>>>> ---
>>>>> tools/lib/bpf/xsk.c | 8 ++------
>>>>> 1 file changed, 2 insertions(+), 6 deletions(-)
>>>>>
>>>>> diff --git a/tools/lib/bpf/xsk.c b/tools/lib/bpf/xsk.c
>>>>> index 680e63066cf3..598e487d9ce8 100644
>>>>> --- a/tools/lib/bpf/xsk.c
>>>>> +++ b/tools/lib/bpf/xsk.c
>>>>> @@ -603,12 +603,8 @@ int xsk_socket__create(struct xsk_socket **xsk_ptr, const char *ifname,
>>>>>
>>>>> optlen = sizeof(opts);
>>>>> err = getsockopt(xsk->fd, SOL_XDP, XDP_OPTIONS, &opts, &optlen);
>>>>> - if (err) {
>>>>> - err = -errno;
>>>>> - goto out_mmap_tx;
>>>>> - }
>>>>> -
>>>>> - xsk->zc = opts.flags & XDP_OPTIONS_ZEROCOPY;
>>>>> + if (!err)
>>>>> + xsk->zc = opts.flags & XDP_OPTIONS_ZEROCOPY;
>>>>>
>>>>> if (!(xsk->config.libbpf_flags & XSK_LIBBPF_FLAGS__INHIBIT_PROG_LOAD)) {
>>>>> err = xsk_setup_xdp_prog(xsk);
>>>>
>>>> Since 'zc' is not used by anybody, maybe all codes 'zc' related can be
>>>> removed? It can be added back back once there is an interface to use
>>>> 'zc'?
>>>
>>> Fine with me; up to the maintainers what they prefer, I guess? :)
>
> Given this is not exposed to applications at this point and we don't do anything
> useful with it, lets just remove the zc cruft until there is a proper interface
> added to libbpf. Toke, please respin with the suggested removal, thanks!
Sure, can do :)
-Toke
^ permalink raw reply
* Re: [PATCH net] net/sched: cls_api: Fix nooffloaddevcnt counter in indr block call success
From: wenxu @ 2019-09-16 13:28 UTC (permalink / raw)
To: Jiri Pirko; +Cc: davem, netdev
In-Reply-To: <20190916102810.GN2286@nanopsycho.orion>
在 2019/9/16 18:28, Jiri Pirko 写道:
> Please use get_maintainers script to get list of ccs.
>
> Mon, Sep 16, 2019 at 12:15:34PM CEST, wenxu@ucloud.cn wrote:
>> From: wenxu <wenxu@ucloud.cn>
>>
>> When a block bind with a dev which support indr block call(vxlan/gretap
>> device). It can bind success but with nooffloaddevcnt++. It will fail
>> when replace the hw filter in tc_setup_cb_call with skip_sw mode for
>> checkout the nooffloaddevcnt and skip_sw.
> I read this paragraph 5 times, I still don't understand :( Could you
> please re-phrase?
will do.
>> if (block->nooffloaddevcnt && err_stop)
>> return -EOPNOTSUPP;
>>
>> So with this patch, if the indr block call success, it will not modify
>> the nooffloaddevcnt counter.
>>
>> Fixes: 7f76fa36754b ("net: sched: register callbacks for indirect tc block binds")
>> Signed-off-by: wenxu <wenxu@ucloud.cn>
>> ---
>> net/sched/cls_api.c | 27 +++++++++++++++------------
>> 1 file changed, 15 insertions(+), 12 deletions(-)
>>
>> diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c
>> index efd3cfb..8a1e3a5 100644
>> --- a/net/sched/cls_api.c
>> +++ b/net/sched/cls_api.c
>> @@ -766,10 +766,10 @@ void tc_indr_block_cb_unregister(struct net_device *dev,
>> }
>> EXPORT_SYMBOL_GPL(tc_indr_block_cb_unregister);
>>
>> -static void tc_indr_block_call(struct tcf_block *block, struct net_device *dev,
>> - struct tcf_block_ext_info *ei,
>> - enum flow_block_command command,
>> - struct netlink_ext_ack *extack)
>> +static int tc_indr_block_call(struct tcf_block *block, struct net_device *dev,
>> + struct tcf_block_ext_info *ei,
>> + enum flow_block_command command,
>> + struct netlink_ext_ack *extack)
>> {
>> struct tc_indr_block_cb *indr_block_cb;
>> struct tc_indr_block_dev *indr_dev;
>> @@ -785,7 +785,7 @@ static void tc_indr_block_call(struct tcf_block *block, struct net_device *dev,
>>
>> indr_dev = tc_indr_block_dev_lookup(dev);
>> if (!indr_dev)
>> - return;
>> + return -ENOENT;
>>
>> indr_dev->block = command == FLOW_BLOCK_BIND ? block : NULL;
>>
>> @@ -793,7 +793,10 @@ static void tc_indr_block_call(struct tcf_block *block, struct net_device *dev,
>> indr_block_cb->cb(dev, indr_block_cb->cb_priv, TC_SETUP_BLOCK,
>> &bo);
>>
>> - tcf_block_setup(block, &bo);
>> + if (list_empty(&bo.cb_list))
>> + return -EOPNOTSUPP;
> How is this part related to the rest of the patch?
This check the list is to make sure there is real hw-offload for the indr device( There maybe
get some wrong with indr_block_cb->cb in the driver). It should a return err to indicate the
caller add the nooffloaddevcnt counter.
.
^ 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