Currently, at xdp_adjust_tail_kern.c, MAX_PCKT_SIZE is limited
to 600. To make this size flexible, static global variable
'max_pcktsz' is added.
By updating new packet size from the user space, xdp_adjust_tail_kern.o
will use this value as a new max packet size.
This static global variable can be
From: Sameeh Jubran
Since we use the same IRQ and NAPI to service RX and TX then we need to
use a combined channel instead of rx and tx channels.
Signed-off-by: Sameeh Jubran
---
drivers/net/ethernet/amazon/ena/ena_ethtool.c | 10 ++
1 file changed, 2 insertions(+), 8 deletions(-)
dif
From: Sameeh Jubran
The number of queues can be derived using ethtool, no need to print
it in ena_probe()
Signed-off-by: Sameeh Jubran
---
drivers/net/ethernet/amazon/ena/ena_netdev.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ethernet/amazon/ena/ena_ne
From: Sameeh Jubran
Most places in the code refer to the IO queues as io_queues and not
simply queues. Examples - max_io_queues_per_vf, ENA_MAX_NUM_IO_QUEUES,
ena_destroy_all_io_queues() etc..
We are also adding the new max_num_io_queues field to struct ena_adapter
in the following commit.
The
From: Sameeh Jubran
Difference from v2:
* ethtool's set/get channels: Switched to using combined instead of
separate rx/tx
* Fixed error handling in set_channels
* Fixed indentation and cosmetic issues as requested by Jakub Kicinski
Difference from v1:
* Dropped the print from patch 0002 - "n
From: Sameeh Jubran
- Rename ena_calc_queue_size() to ena_calc_io_queue_size() for clarity
and consistency
- Remove redundant number of io queues parameter in functions
ena_enable_msix() and ena_enable_msix_and_set_admin_interrupts(),
which already get adapter parameter, so use adapter->num
From: Sameeh Jubran
- Update ena_ethtool:ena_get_channels() to return adapter->max_io_queues
so that ethtool -l returns the correct maximum queue number.
- Change the name of ena_calc_io_queue_num() to
ena_calc_max_io_queue_num() as it returns the maximum number of io
queues and actual num
From: Sameeh Jubran
Set channels callback enables the user to change the count of queues
used by the driver using ethtool. We decided to currently support only
equal number of rx and tx queues, this might change in the future.
Also rename dev_up to dev_was_up in ena_update_queue_count() to make
On Thu, Oct 03, 2019 at 02:57:59PM -0700, Tom Herbert wrote:
> From: Tom Herbert
>
> Move generic functions in exthdrs.c to new exthdrs_common.c so that
> exthdrs.c only contains functions that are specific to IPv6 processing,
> and exthdrs_common.c contains functions that are generic. These
> fu
On Thu, Oct 03, 2019 at 02:57:58PM -0700, Tom Herbert wrote:
> From: Tom Herbert
>
> Create exthdrs_options.c to hold code related to specific Hop-by-Hop
> and Destination extension header options. Move related functions in
> exthdrs.c to the new file.
>
> Create include net/ipeh.h to contain co
On Thu, Oct 03, 2019 at 02:58:00PM -0700, Tom Herbert wrote:
> From: Tom Herbert
>
> Create a generic TLV parser. This will be used with various
> extension headers that carry options including Destination,
> Hop-by-Hop, Segment Routing TLVs, and other cases of simple
> stateless parsing.
>
> Si
On Thu, Oct 03, 2019 at 02:58:01PM -0700, Tom Herbert wrote:
> From: Tom Herbert
>
> Create a single TLV parameter table that holds meta information for IPv6
> Hop-by-Hop and Destination TLVs. The data structure is composed of a 256
> element array of u8's (one entry for each TLV type to allow O(
From: Yi-Hung Wei
Date: Fri, 4 Oct 2019 09:26:44 -0700
> This patch allows to attach conntrack helper to a confirmed conntrack
> entry. Currently, we can only attach alg helper to a conntrack entry
> when it is in the unconfirmed state. This patch enables an use case
> that we can firstly comm
On Thu, Oct 03, 2019 at 02:58:02PM -0700, Tom Herbert wrote:
> From: Tom Herbert
>
> Define a number of transmit parameters for TLV Parameter table
> definitions. These will be used for validating TLVs that are set
> on a socket.
>
> Signed-off-by: Tom Herbert
> ---
> include/net/ipeh.h
From: "Daniel T. Lee"
Date: Sat, 5 Oct 2019 17:25:05 +0900
> Currently, pktgen script supports specify destination port range.
>
> To further extend the capabilities, this commit allows to specify destination
> IP range with CIDR when running pktgen script.
>
> Specifying destination IP range
From: Jiri Pirko
Date: Sat, 5 Oct 2019 20:04:32 +0200
> From: Jiri Pirko
>
> In generic netlink, parsing attributes for doit() callback is already
> implemented. They are available in info->attrs.
>
> For dumpit() however, each user which is interested in attributes have to
> parse it manuall
Jakub Kicinski writes:
> On Sat, 05 Oct 2019 12:29:14 +0200, Toke Høiland-Jørgensen wrote:
>> >> +static int bpf_inject_chain_calls(struct bpf_verifier_env *env)
>> >> +{
>> >> + struct bpf_prog *prog = env->prog;
>> >> + struct bpf_insn *insn = prog->insnsi;
>> >> + int i, cnt, delta = 0, ret =
From: Yamin Friedman
Expose maximum scatter entries per RDMA READ for optimal performance.
Signed-off-by: Yamin Friedman
Reviewed-by: Or Gerlitz
Signed-off-by: Leon Romanovsky
---
include/linux/mlx5/mlx5_ifc.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/include/linux/
From: Leon Romanovsky
Hi,
This series from Yamin implements long standing "TODO" existed in rw.c.
Thanks
Yamin Friedman (3):
net/mlx5: Expose optimal performance scatter entries capability
RDMA/mlx5: Add capability for max sge to get optimized performance
RDMA/rw: Support threshold for r
From: Yamin Friedman
Allows the IB device to provide a value of maximum scatter gather entries
per RDMA READ.
In certain cases it may be preferable for a device to perform UMR memory
registration rather than have many scatter entries in a single RDMA READ.
This provides a significant performance
From: Yamin Friedman
If there are more scatter entries than the recommended limit provided by
the ib device, UMR registration is used. This will provide optimal
performance when performing large RDMA READs over devices that advertise
the threshold capability.
With ConnectX-5 running NVMeoF RDMA
This function de-facto returns a bool, so let's change the return type
accordingly.
Signed-off-by: Heiner Kallweit
---
include/linux/skbuff.h | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 4351577b1..0a58402a1 10064
From: Jiri Pirko
Date: Sun, 6 Oct 2019 08:30:00 +0200
> From: Jiri Pirko
>
> Allow user to test devlink reload failures: Fail to reload and fail
> during reload.
Series applied.
From: Ido Schimmel
Date: Sun, 6 Oct 2019 09:34:47 +0300
> From: Ido Schimmel
>
> Vadim says:
>
> The patchset adds support for a new field "num_of_modules" of Management
> General Peripheral Information Register (MGPIR), providing the maximum
> number of QSFP modules, which can be supported b
Use helper skb_ensure_writable in two more places to simplify the code.
Signed-off-by: Heiner Kallweit
---
net/core/dev.c | 20
1 file changed, 8 insertions(+), 12 deletions(-)
diff --git a/net/core/dev.c b/net/core/dev.c
index 944de67ee..7d05e042c 100644
--- a/net/core/dev
This got converted from strncpy to strlcpy but it's
now not necessary to use one character less than the
actual size.
Perhaps the sizeof() - 1 is now not correct and it
should use strscpy and a normal sizeof.
from:
commit 7eb74ff891b4e94b8bac48f648a21e4b94ddee64
Author: Mitch Williams
Date: M
When the netdev is down, the queues and their debug stats
do not exist, so don't try using a pointer to them when
when printing the ethtool stats.
Fixes: e470355bd96a ("ionic: Add driver stats")
Signed-off-by: Shannon Nelson
---
.../net/ethernet/pensando/ionic/ionic_lif.h | 2 ++
.../net/ethe
Hello,
Patches 1-2 fix the memory leaks that syzbot has reported in net/llc:
memory leak in llc_ui_create (2)
memory leak in llc_ui_sendmsg
memory leak in llc_conn_ac_send_sabme_cmd_p_set_x
Patches 3-4 fix related bugs that I noticed while reading this code.
Note: I've t
From: Eric Biggers
syzbot reported:
BUG: memory leak
unreferenced object 0x888116270800 (size 224):
comm "syz-executor641", pid 7047, jiffies 4294947360 (age 13.860s)
hex dump (first 32 bytes):
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
From: Eric Biggers
If llc_conn_state_process() sees that llc_conn_service() put the skb on
a list, it will drop one fewer references to it. This is wrong because
the current behavior is that llc_conn_service() never consumes a
reference to the skb.
The code also makes the number of skb referenc
From: Eric Biggers
syzbot reported:
BUG: memory leak
unreferenced object 0x88811eb3de00 (size 224):
comm "syz-executor559", pid 7315, jiffies 4294943019 (age 10.300s)
hex dump (first 32 bytes):
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
From: Eric Biggers
All callers of llc_conn_state_process() except llc_build_and_send_pkt()
(via llc_ui_sendmsg() -> llc_ui_send_data()) assume that it always
consumes a reference to the skb. Fix this caller to do the same.
Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Signed-off-by: Eric Biggers
--
On Sat, Oct 05, 2019 at 11:36:16PM -0700, Andrii Nakryiko wrote:
> On Fri, Oct 4, 2019 at 10:08 PM Alexei Starovoitov wrote:
> >
> > If in-kernel BTF exists parse it and prepare 'struct btf *btf_vmlinux'
> > for further use by the verifier.
> > In-kernel BTF is trusted just like kallsyms and other
On Sat, Oct 5, 2019 at 10:46 PM Andrii Nakryiko wrote:
>
> Get rid of list of BPF helpers in bpf_helpers.h (irony...) and
> auto-generate it into bpf_helpers_defs.h, which is now included from
> bpf_helpers.h.
>
> Suggested-by: Alexei Starovoitov
> Signed-off-by: Andrii Nakryiko
> ---
> tools/l
On Sun, Oct 6, 2019 at 4:56 PM Alexei Starovoitov
wrote:
>
> On Sat, Oct 5, 2019 at 10:46 PM Andrii Nakryiko wrote:
> >
> > Get rid of list of BPF helpers in bpf_helpers.h (irony...) and
> > auto-generate it into bpf_helpers_defs.h, which is now included from
> > bpf_helpers.h.
> >
> > Suggested-
On Sun, Oct 6, 2019 at 4:49 PM Alexei Starovoitov
wrote:
>
> On Sat, Oct 05, 2019 at 11:36:16PM -0700, Andrii Nakryiko wrote:
> > On Fri, Oct 4, 2019 at 10:08 PM Alexei Starovoitov wrote:
> > >
> > > If in-kernel BTF exists parse it and prepare 'struct btf *btf_vmlinux'
> > > for further use by t
On Fri, Oct 04, 2019 at 07:22:41PM +0200, Toke Høiland-Jørgensen wrote:
> From: Toke Høiland-Jørgensen
>
> This adds support for injecting chain call logic into eBPF programs before
> they return. The code injection is controlled by a flag at program load
> time; if the flag is set, the verifier
On Sun, Oct 6, 2019 at 5:13 PM Andrii Nakryiko
wrote:
>
> On Sun, Oct 6, 2019 at 4:56 PM Alexei Starovoitov
> wrote:
> >
> > On Sat, Oct 5, 2019 at 10:46 PM Andrii Nakryiko wrote:
> > >
> > > Get rid of list of BPF helpers in bpf_helpers.h (irony...) and
> > > auto-generate it into bpf_helpers_d
On Sun, Oct 6, 2019 at 5:32 PM Alexei Starovoitov
wrote:
>
> On Sun, Oct 6, 2019 at 5:13 PM Andrii Nakryiko
> wrote:
> >
> > On Sun, Oct 6, 2019 at 4:56 PM Alexei Starovoitov
> > wrote:
> > >
> > > On Sat, Oct 5, 2019 at 10:46 PM Andrii Nakryiko wrote:
> > > >
> > > > Get rid of list of BPF hel
Get rid of list of BPF helpers in bpf_helpers.h (irony...) and
auto-generate it into bpf_helpers_defs.h, which is now included from
bpf_helpers.h.
Suggested-by: Alexei Starovoitov
Signed-off-by: Andrii Nakryiko
---
tools/testing/selftests/bpf/.gitignore| 1 +
tools/testing/selftests/bpf/M
Enhance scripts/bpf_helpers_doc.py to emit C header with BPF helper
definitions (to be included from libbpf's bpf_helpers.h).
Signed-off-by: Andrii Nakryiko
---
scripts/bpf_helpers_doc.py | 155 -
1 file changed, 154 insertions(+), 1 deletion(-)
diff --git a/
Various small fixes to BPF helper documentation comments, enabling
automatic header generation with a list of BPF helpers.
Signed-off-by: Andrii Nakryiko
---
include/uapi/linux/bpf.h | 32
tools/include/uapi/linux/bpf.h | 32
This patch set adds ability to auto-generate list of BPF helper definitions.
It relies on existing scripts/bpf_helpers_doc.py and include/uapi/linux/bpf.h
having a well-defined set of comments. bpf_helper_defs.h contains all BPF
helper signatures which stay in sync with latest bpf.h UAPI. This
auto
On Sun, Oct 6, 2019 at 7:24 PM Andrii Nakryiko
wrote:
>
> On Sun, Oct 6, 2019 at 5:32 PM Alexei Starovoitov
> wrote:
> >
> > On Sun, Oct 6, 2019 at 5:13 PM Andrii Nakryiko
> > wrote:
> > >
> > > On Sun, Oct 6, 2019 at 4:56 PM Alexei Starovoitov
> > > wrote:
> > > >
> > > > On Sat, Oct 5, 2019 a
Current Makefile dependency chain is not strict enough and allows
test_attach_probe.o to be built before test_progs's
prog_test/attach_probe.o is built, which leads to assembler compainig
about missing included binary.
This patch is a minimal fix to fix this issue by enforcing that
test_attach_pro
On Sun, Oct 6, 2019 at 4:58 AM Daniel T. Lee wrote:
>
> Currently, at xdp_adjust_tail_kern.c, MAX_PCKT_SIZE is limited
> to 600. To make this size flexible, static global variable
> 'max_pcktsz' is added.
>
> By updating new packet size from the user space, xdp_adjust_tail_kern.o
> will use this v
TLS requires the entire record to fit into the receive buffer.
If user sets the receive buffer too low when larger record
arrives strparser returns EMSGSIZE.
Add a test case for this situation.
Signed-off-by: Jakub Kicinski
Reviewed-by: Simon Horman
---
tools/testing/selftests/net/tls.c | 28 +
If strparser encounters an error it reports it on the socket and
stops any further processing. TLS RX will currently pick up that
error code with sock_error(), and report it to user space once.
Subsequent read calls will block indefinitely.
Since the error condition is not cleared and processing i
Hi!
I was investigating the correct behaviour around small receive
window size and were slightly surprised to find that currently,
if the recv buffer is too small to fit the entire record, TLS will
report -EMSGSIZE once, and then block indefinitely on a subsequent
call.
Perhaps naively I'd think
Store async_capable on a single bit instead of a full integer
to save space.
Signed-off-by: Jakub Kicinski
Reviewed-by: Simon Horman
---
include/net/tls.h | 4 ++--
net/tls/tls_sw.c | 5 +++--
2 files changed, 5 insertions(+), 4 deletions(-)
diff --git a/include/net/tls.h b/include/net/tls.h
Hi!
This set brings a number of minor code changes from my tree which
don't have a noticeable impact on performance but seem reasonable
nonetheless.
First sk_msg_sg copy array is converted to a bitmap, zeroing that
structure takes a lot of time, hence we should try to keep it
small.
Next two con
Don't use bool array in struct sk_msg_sg, save 12 bytes.
Signed-off-by: Jakub Kicinski
Reviewed-by: Dirk van der Merwe
---
include/linux/skmsg.h | 12
net/core/filter.c | 4 ++--
2 files changed, 10 insertions(+), 6 deletions(-)
diff --git a/include/linux/skmsg.h b/include/li
Tell GCC sk->err is not likely to be set.
Signed-off-by: Jakub Kicinski
Reviewed-by: Dirk van der Merwe
---
net/tls/tls_device.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/net/tls/tls_device.c b/net/tls/tls_device.c
index f306e4c7bf15..fcf38edc07d6 100644
--- a/net/tls/
Use a single bit instead of boolean to remember if packet
was already decrypted.
Signed-off-by: Jakub Kicinski
Reviewed-by: Simon Horman
---
include/net/tls.h | 2 +-
net/tls/tls_sw.c | 6 +++---
2 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/include/net/tls.h b/include/net/tls
Make sure GCC realizes it's unlikely that allocations will fail.
Signed-off-by: Jakub Kicinski
Reviewed-by: Dirk van der Merwe
---
net/tls/tls_device.c | 5 ++---
1 file changed, 2 insertions(+), 3 deletions(-)
diff --git a/net/tls/tls_device.c b/net/tls/tls_device.c
index fcf38edc07d6..23c19b
Avoid unnecessary pointer chasing and calculations, callers already
have most of the state tls_device_decrypted() needs.
Signed-off-by: Jakub Kicinski
Reviewed-by: Dirk van der Merwe
---
include/net/tls.h| 7 +--
net/tls/tls_device.c | 5 ++---
net/tls/tls_sw.c | 2 +-
3 files chang
raw_tracepoints are an eBPF alternative to standard tracepoints which
attach to a tracepoint without the perf layer being executed, making
them faster.
Since xdp_monitor is supposed to have as little impact on the system as
possible it is switched to using them by append raw_ to the SEC names.
Th
On Sun, Oct 6, 2019 at 8:31 PM Andrii Nakryiko wrote:
>
> Current Makefile dependency chain is not strict enough and allows
> test_attach_probe.o to be built before test_progs's
> prog_test/attach_probe.o is built, which leads to assembler compainig
> about missing included binary.
>
> This patch
On Sun, Oct 6, 2019 at 8:08 PM Andrii Nakryiko wrote:
>
> This patch set adds ability to auto-generate list of BPF helper definitions.
> It relies on existing scripts/bpf_helpers_doc.py and include/uapi/linux/bpf.h
> having a well-defined set of comments. bpf_helper_defs.h contains all BPF
> helpe
> /*
> - * Check if the device might use memory registration. This is currently only
> - * true for iWarp devices. In the future we can hopefully fine tune this
> based
> - * on HCA driver input.
> + * Check if the device might use memory registration.
> */
Please keep the important bits of t
60 matches
Mail list logo