Ensure that failure on any individual test results in an overall
failure of the test script.
Signed-off-by: Fred Klassen
---
tools/testing/selftests/net/udpgso_bench.sh | 33 +
1 file changed, 33 insertions(+)
diff --git a/tools/testing/selftests/net/udpgso_bench.sh
: Unexpected number of Zerocopy completions:493706
expected493704 received
Also reduce individual test time from 4 to 3 seconds so that
overall test time does not increase significantly.
v3: Enhancements as per Willem de Bruijn
- document -P option for TCP audit
Signed-off-by: Fred
MSG_TRUNC
- general code cleanup
Signed-off-by: Fred Klassen
---
tools/testing/selftests/net/udpgso_bench_tx.c | 291 --
1 file changed, 276 insertions(+), 15 deletions(-)
diff --git a/tools/testing/selftests/net/udpgso_bench_tx.c
b/tools/testing/selftests/net
Updates to UDP GSO selftests ot optionally stress test CMSG
subsytem, and report the reliability and performance of both
TX Timestamping and ZEROCOPY messages.
Fred Klassen (3):
net/udpgso_bench_tx: options to exercise TX CMSG
net/udpgso_bench.sh add UDP GSO audit tests
net/udpgso_bench.sh
tests as noted by Willem de Bruijn
Moving tests from net to net-next
v3: Update only relevant tx_flag bits as per
Willem de Bruijn
v4: Update comments and commit message as per
Willem de Bruijn
Fixes: ee80d1ebe5ba ("udp: add udp gso")
Signed-off-by: Fred Klassen
---
net/ip
> On May 28, 2019, at 2:35 PM, Willem de Bruijn
> wrote:
>> -static void flush_zerocopy(int fd)
>> +static void flush_cmsg(struct cmsghdr *cmsg)
>> {
>> - struct msghdr msg = {0};/* flush */
>> + switch (cmsg->cmsg_level) {
>> + case SOL_SOCKET:
>> + if (cm
Ensure that failure on any individual test results in an overall
failure of the test script.
Signed-off-by: Fred Klassen
---
tools/testing/selftests/net/udpgso_bench.sh | 33 +
1 file changed, 33 insertions(+)
diff --git a/tools/testing/selftests/net/udpgso_bench.sh
)
./udpgso_bench_tx: Unexpected number of Zerocopy completions:493706
expected493704 received
Also reduce individual test time from 4 to 3 seconds so that
overall test time does not increase significantly.
Signed-off-by: Fred Klassen
---
tools/testing/selftests/net/udpgso_bench.sh | 18
: Enhancements as per Willem de Bruijn
- Updated control and buffer parameters for recvmsg
- poll() parameter cleanup
- fail on bad audit results
- remove TOS options
- improved reporting
Signed-off-by: Fred Klassen
---
tools/testing/selftests/net/udpgso_bench_tx.c | 324
Updates to UDP GSO selftests ot optionally stress test CMSG
subsytem, and report the reliability and performance of both
TX Timestamping and ZEROCOPY messages.
Fred Klassen (3):
net/udpgso_bench_tx: options to exercise TX CMSG
net/udpgso_bench.sh add UDP GSO audit tests
net/udpgso_bench.sh
Fixes an issue where TX Timestamps are not arriving on the error queue
when UDP_SEGMENT CMSG type is combined with CMSG type SO_TIMESTAMPING.
Fred Klassen (1):
net/udp_gso: Allow TX timestamp with UDP GSO
net/ipv4/udp_offload.c | 5 +
1 file changed, 5 insertions(+)
--
2.11.0
nly relevant tx_flag bits as per
Willem de Bruijn
Fixes: ee80d1ebe5ba ("udp: add udp gso")
Signed-off-by: Fred Klassen
---
net/ipv4/udp_offload.c | 5 +
1 file changed, 5 insertions(+)
diff --git a/net/ipv4/udp_offload.c b/net/ipv4/udp_offload.c
index 065334b41d57..de8e
> On May 28, 2019, at 8:08 AM, Willem de Bruijn
> wrote:
>
I will push up latest patches soon.
I did some testing and discovered that only TCP audit tests failed. They
failed much less often when enabling poll. Once in about 20 runs
still failed. Therefore I commented out the TCP audit tes
> On May 27, 2019, at 6:15 PM, Willem de Bruijn
> wrote:
>> I wanted to discuss whether or not to attach a buffer to the
>> recvmsg(fd, &msg, MSG_ERRQUEUE). Without it, I have
>> MSG_TRUNC errors in my msg_flags. Either I have to add
>> a buffer, or ignore that error flag.
>
> Either sounds r
> On May 27, 2019, at 2:46 PM, Willem de Bruijn
> wrote:
>> Also, I my v2 fix in net is still up for debate. In its current state, it
>> meets my application’s requirements, but may not meet all of yours.
> I gave more specific feedback on issues with it (referencing zerocopy
> and IP_TOS, sa
Willem,
Thanks for spending so much time with me on this patch. I’m pretty
new to this, so I know I am making lots of mistakes. I have been
working on a v2 of the selftests in net-next, but want to review the
layout of the report before I submit (see below).
Also, I my v2 fix in net is still up f
Fixes an issue where TX Timestamps are not arriving on the error queue
when UDP_SEGMENT CMSG type is combined with CMSG type SO_TIMESTAMPING.
This is version 2 which moves the tests to net-next.
Fred Klassen (1):
net/udp_gso: Allow TX timestamp with UDP GSO
net/ipv4/udp_offload.c | 4
1
to set SOL_IP/IP_TOS.
./udpgso_bench_tx -4ucPv -S 1472 -q 182 -l2 -D 172.16.120.18
poll timeout
udp tx: 0 MB/s1 calls/s 1 msg/s
This patch preserves tx_flags for the first UDP GSO segment. This
mirrors the stack's behaviour for IPv4 fragments.
Fixes: ee80d1ebe5ba ("udp: add
> On May 23, 2019, at 2:59 PM, Willem de Bruijn
> wrote:what exactly is the issue with IP_TOS?
>
> If I understand correctly, the issue here is that the new 'P' option
> that polls on the error queue times out. This is unrelated to
> specifying TOS bits? Without zerocopy or timestamps, no mes
> On May 23, 2019, at 2:39 PM, Willem de Bruijn
> wrote:
> Zerocopy notification reference count is managed in skb_segment. That
> should work.
>
I’m trying to understand the context of reference counting in skb_segment. I
assume that
there is an opportunity to optimize the count of outstan
> On May 23, 2019, at 2:59 PM, Willem de Bruijn
> wrote:
> what exactly is the issue with IP_TOS?
>
> If I understand correctly, the issue here is that the new 'P' option
> that polls on the error queue times out. This is unrelated to
> specifying TOS bits? Without zerocopy or timestamps, no
> On May 25, 2019, at 8:20 AM, Willem de Bruijn
> wrote:
>
> On Fri, May 24, 2019 at 6:01 PM Fred Klassen wrote:
>>
>>
>>
>>> On May 24, 2019, at 12:29 PM, Willem de Bruijn
>>> wrote:
>>>
>>> It is the last moment th
> On May 24, 2019, at 12:29 PM, Willem de Bruijn
> wrote:
>
> It is the last moment that a timestamp can be generated for the last
> byte, I don't see how that is "neither the start nor the end of a GSO
> packet”.
My misunderstanding. I thought TCP did last segment timestamping, not
last byt
> Interesting. TCP timestamping takes the opposite choice and does
> timestamp the last byte in the sendmsg request.
>
I have a difficult time with the philosophy of TX timestamping the last
segment. The actual timestamp occurs just before the last segment
is sent. This is neither the start nor
>
> To ensure that we do not regress, when adding options, please consider
> (just a general suggestion, not a strong request for this patch set)
> updating the kselftest to run a variant of the test with the new code
> coverage. In this case, make the code pass/fail instead of only user
> interpr
> Thanks for the report.
>
> Zerocopy notification reference count is managed in skb_segment. That
> should work.
>
> Support for timestamping with the new GSO feature is indeed an
> oversight. The solution is similar to how TCP associates the timestamp
> with the right segment in tcp_gso_tstamp.
Willem, this is only my 2nd patch, and my last one was a one liner.
I’ll try to work through this, but let me know if I am doing a rookie
mistake (learning curve and all).
> On May 23, 2019, at 2:56 PM, Willem de Bruijn
> wrote:
>
> On Thu, May 23, 2019 at 5:11 PM Fred Kl
("selftests: udp gso benchmark")
Signed-off-by: Fred Klassen
---
tools/testing/selftests/net/udpgso_bench_tx.c | 290 --
1 file changed, 273 insertions(+), 17 deletions(-)
diff --git a/tools/testing/selftests/net/udpgso_bench_tx.c
b/tools/testing/selftests/net/udpgso_
gs (10885/s)
Tx Timestamps: received: 34046 errors: 0
Zerocopy acks: received: 54422 errors: 0
Fixes: 3a687bef148d ("selftests: udp gso benchmark")
Signed-off-by: Fred Klassen
---
tools/testing/selftests/net/udpgso_bench_tx.c | 152 +++---
1 file
Fixes the following error if using both -m and -u options:
# ./udpgso_bench -4um -D 172.16.120.189
./udpgso_bench: sendmmsg: Destination address required
Fixes: 3a687bef148d ("selftests: udp gso benchmark")
Signed-off-by: Fred Klassen
---
tools/testing/selftests/net/udpgso_
ZEROCOPY messages.
Fred Klassen (4):
net/udp_gso: Allow TX timestamp with UDP GSO
net/udpgso_bench_tx: options to exercise TX CMSG
net/udpgso_bench_tx: fix sendmmsg on unconnected socket
net/udpgso_bench_tx: audit error queue
net/ipv4/udp_offload.c| 4 +
tools
to set SOL_IP/IP_TOS.
./udpgso_bench_tx -4ucPv -S 1472 -q 182 -l2 -D 172.16.120.18
poll timeout
udp tx: 0 MB/s1 calls/s 1 msg/s
This patch preserves tx_flags for the first UDP GSO segment. This
mirrors the stack's behaviour for IPv4 fragments.
Fixes: ee80d1ebe5ba ("udp: add
Correct, targeting 5.0.
> On Jan 19, 2019, at 5:30 PM, Corey Minyard wrote:
>
> I assume you want this for 5.0., right?
handle_new_recv_msgs+0x1ce/0x440
tasklet_action_common.isra.19+0xc4/0x250
__do_softirq+0x11f/0x51f
Fixes: e86ee2d44b44 ("ipmi: Rework locking and shutdown for hot remove")
Cc: sta...@vger.kernel.org
Signed-off-by: Fred Klassen
---
drivers/char/ipmi/ipmi_msghandler.c | 2 +-
1 file changed, 1 inser
/0x21d0
handle_new_recv_msgs+0x1ce/0x440
tasklet_action_common.isra.19+0xc4/0x250
__do_softirq+0x11f/0x51f
Fixes: e86ee2d44b4 ("ipmi: Rework locking and shutdown for hot remove")
Signed-off-by: Fred Klassen
Cc: sta...@vger.kernel.org
---
drivers/char/ipmi/ipmi_msghandler.c | 2 +-
1 fi
handle_one_recv_msg+0x37a/0x21d0
handle_new_recv_msgs+0x1ce/0x440
tasklet_action_common.isra.19+0xc4/0x250
__do_softirq+0x11f/0x51f
Fixes: e86ee2d44b4 ("ipmi: Rework locking and shutdown for hot remove")
Signed-off-by: Fred Klassen
---
drivers/char/ipmi/ipmi_msghandler.c | 2
36 matches
Mail list logo