6 messages, 4 authors, 2017-01-25 · open the first message on its own page
From: David Miller <davem@davemloft.net>
Date: 2017-01-25 17:22:11
From: Wei Wang <redacted> Date: Wed, 25 Jan 2017 09:15:34 -0800
Looks like you sent a separate patch on top of this patch series to address double connect(). Then I think this patch series should be good to go.
Indeed, Willy please give some kind of ACK. Thanks.
From: Willy Tarreau <w@1wt.eu>
Date: 2017-01-25 17:55:05
On Wed, Jan 25, 2017 at 12:22:05PM -0500, David Miller wrote:
From: Wei Wang <redacted> Date: Wed, 25 Jan 2017 09:15:34 -0800quoted
Looks like you sent a separate patch on top of this patch series to address double connect(). Then I think this patch series should be good to go.Indeed, Willy please give some kind of ACK.
Yes sorry David, for me it's OK. If Wei runs his whole series of tests on it again, it should be perfect. thanks, Willy
From: Wei Wang <hidden>
Date: 2017-01-25 18:55:16
Yes sorry David, for me it's OK. If Wei runs his whole series of tests on it again, it should be perfect.
I just ran all the TFO related tests with Willy's patch on top of this patch series. And everything passes. On Wed, Jan 25, 2017 at 9:54 AM, Willy Tarreau [off-list ref] wrote:
On Wed, Jan 25, 2017 at 12:22:05PM -0500, David Miller wrote:quoted
From: Wei Wang <redacted> Date: Wed, 25 Jan 2017 09:15:34 -0800quoted
Looks like you sent a separate patch on top of this patch series to address double connect(). Then I think this patch series should be good to go.Indeed, Willy please give some kind of ACK.Yes sorry David, for me it's OK. If Wei runs his whole series of tests on it again, it should be perfect. thanks, Willy
From: Eric Dumazet <hidden>
Date: 2017-01-25 19:03:06
On Wed, 2017-01-25 at 10:54 -0800, Wei Wang wrote:
quoted
Yes sorry David, for me it's OK. If Wei runs his whole series of tests on it again, it should be perfect.I just ran all the TFO related tests with Willy's patch on top of this patch series. And everything passes.
Note that I am not sure we correctly test that a second connect() can be done, and I am not sure kernel would check that the remote IP and destination port is the same. Ie what should happen for setsockopt(fd, SOL_TCP, TCP_FASTOPEN_CONNECT, &on, 4) connect(fd, "1.2.3.4:80") connect(fd, "55.66.77.88:4000") This multiple connect() thing should have been forbidden in the first place really.
From: David Miller <davem@davemloft.net>
Date: 2017-01-25 19:03:35
From: Wei Wang <redacted> Date: Wed, 25 Jan 2017 10:54:50 -0800
quoted
Yes sorry David, for me it's OK. If Wei runs his whole series of tests on it again, it should be perfect.I just ran all the TFO related tests with Willy's patch on top of this patch series. And everything passes.
Great, I'll apply everything, thanks.
From: Wei Wang <hidden>
Date: 2017-01-25 19:30:07
Note that I am not sure we correctly test that a second connect() can be done, and I am not sure kernel would check that the remote IP and destination port is the same.
Ie what should happen for
setsockopt(fd, SOL_TCP, TCP_FASTOPEN_CONNECT, &on, 4) connect(fd, "1.2.3.4:80") connect(fd, "55.66.77.88:4000")
I wrote a simple code to test this scenario and the second connect() returns EISCONN as well even though the destination IP is different. On Wed, Jan 25, 2017 at 11:03 AM, David Miller [off-list ref] wrote:
From: Wei Wang <redacted> Date: Wed, 25 Jan 2017 10:54:50 -0800quoted
quoted
Yes sorry David, for me it's OK. If Wei runs his whole series of tests on it again, it should be perfect.I just ran all the TFO related tests with Willy's patch on top of this patch series. And everything passes.Great, I'll apply everything, thanks.