* [dpdk-dev] e1000: randomly loosing link change events triggered by the peer @ 2016-03-26 1:42 Marc Sune 2016-03-28 1:54 ` Lu, Wenzhuo 0 siblings, 1 reply; 5+ messages in thread From: Marc Sune @ 2016-03-26 1:42 UTC (permalink / raw) To: dev, Wenzhuo Lu I found that in the current HEAD in master testing it with an I218-LM in autoneg mode, when link is forced to be renegociated by the peer (e.g. via ethtool on a peer Linux box) _some_ change events are lost. It is quite random, but it seems to happen more while changing the speed than when changing the duplex mode. However, when one or more link change events have been lost and the phy medium is disconnected and reconnected, the link speed and duplex mode are then correctly updated. Marc ^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [dpdk-dev] e1000: randomly loosing link change events triggered by the peer 2016-03-26 1:42 [dpdk-dev] e1000: randomly loosing link change events triggered by the peer Marc Sune @ 2016-03-28 1:54 ` Lu, Wenzhuo 2016-03-28 11:02 ` Marc 0 siblings, 1 reply; 5+ messages in thread From: Lu, Wenzhuo @ 2016-03-28 1:54 UTC (permalink / raw) To: Marc Sune, dev Hi Marc From: Marc Sune [mailto:marcdevel@gmail.com] Sent: Saturday, March 26, 2016 9:43 AM To: dev@dpdk.org; Lu, Wenzhuo Subject: e1000: randomly loosing link change events triggered by the peer I found that in the current HEAD in master testing it with an I218-LM in autoneg mode, when link is forced to be renegociated by the peer (e.g. via ethtool on a peer Linux box) _some_ change events are lost. It is quite random, but it seems to happen more while changing the speed than when changing the duplex mode. However, when one or more link change events have been lost and the phy medium is disconnected and reconnected, the link speed and duplex mode are then correctly updated. Marc [Wenzhuo] Thanks for let us know this issue. May I ask some questions? Do you mean this NIC 0x155A? About how to reproduce this problem, you mean use these CLIs, ethtool –s xxx advertise xxx, ethtool –s xxx duplex half/full, to change the peer port’s configuration? ^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [dpdk-dev] e1000: randomly loosing link change events triggered by the peer 2016-03-28 1:54 ` Lu, Wenzhuo @ 2016-03-28 11:02 ` Marc 2016-03-29 1:19 ` Lu, Wenzhuo 0 siblings, 1 reply; 5+ messages in thread From: Marc @ 2016-03-28 11:02 UTC (permalink / raw) To: Lu, Wenzhuo; +Cc: dev On 28 March 2016 at 03:54, Lu, Wenzhuo <wenzhuo.lu@intel.com> wrote: > Hi Marc > > > > *From:* Marc Sune [mailto:marcdevel@gmail.com] > *Sent:* Saturday, March 26, 2016 9:43 AM > *To:* dev@dpdk.org; Lu, Wenzhuo > *Subject:* e1000: randomly loosing link change events triggered by the > peer > > > > I found that in the current HEAD in master testing it with an I218-LM in > autoneg mode, when link is forced to be renegociated by the peer (e.g. via > ethtool on a peer Linux box) _some_ change events are lost. > > > > It is quite random, but it seems to happen more while changing the speed > than when changing the duplex mode. > > > > However, when one or more link change events have been lost and the phy > medium is disconnected and reconnected, the link speed and duplex mode are > then correctly updated. > > > > Marc > > > > [Wenzhuo] Thanks for let us know this issue. May I ask some questions? Do > you mean this NIC 0x155A? > 0x15A2, I218-LM (rev 03) EAL: PCI device 0000:00:19.0 on NUMA socket -1 EAL: probe driver: 8086:15a2 rte_em_pmd EAL: PCI memory mapped at 0x7f85cf400000 EAL: PCI memory mapped at 0x7f85cf420000 PMD: eth_em_dev_init(): port_id 0 vendorID=0x8086 deviceID=0x15a2 I think this is not a NIC issue, but a general problem of the driver (or em code). > About how to reproduce this problem, you mean use these CLIs, ethtool –s > xxx advertise xxx, ethtool –s xxx duplex half/full, to change the peer > port’s configuration? > Correct. I modified l2fwd to check link status and print it on each port stats print iteration. Then from the peer I modified the link properties via ethtool. The result is that transitions from autoneg speeds and/or duplex mode settings are randomly not detected (rte_eth_link in rte_eth_dev_data is not updated), and it prints not-up-to-date state. Marc ^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [dpdk-dev] e1000: randomly loosing link change events triggered by the peer 2016-03-28 11:02 ` Marc @ 2016-03-29 1:19 ` Lu, Wenzhuo 2016-03-30 8:04 ` Marc 0 siblings, 1 reply; 5+ messages in thread From: Lu, Wenzhuo @ 2016-03-29 1:19 UTC (permalink / raw) To: Marc; +Cc: dev Hi Marc, From: marc.sune@gmail.com [mailto:marc.sune@gmail.com] On Behalf Of Marc Sent: Monday, March 28, 2016 7:03 PM To: Lu, Wenzhuo Cc: dev@dpdk.org Subject: Re: e1000: randomly loosing link change events triggered by the peer On 28 March 2016 at 03:54, Lu, Wenzhuo <wenzhuo.lu@intel.com<mailto:wenzhuo.lu@intel.com>> wrote: Hi Marc From: Marc Sune [mailto:marcdevel@gmail.com<mailto:marcdevel@gmail.com>] Sent: Saturday, March 26, 2016 9:43 AM To: dev@dpdk.org<mailto:dev@dpdk.org>; Lu, Wenzhuo Subject: e1000: randomly loosing link change events triggered by the peer I found that in the current HEAD in master testing it with an I218-LM in autoneg mode, when link is forced to be renegociated by the peer (e.g. via ethtool on a peer Linux box) _some_ change events are lost. It is quite random, but it seems to happen more while changing the speed than when changing the duplex mode. However, when one or more link change events have been lost and the phy medium is disconnected and reconnected, the link speed and duplex mode are then correctly updated. Marc [Wenzhuo] Thanks for let us know this issue. May I ask some questions? Do you mean this NIC 0x155A? 0x15A2, I218-LM (rev 03) EAL: PCI device 0000:00:19.0 on NUMA socket -1 EAL: probe driver: 8086:15a2 rte_em_pmd EAL: PCI memory mapped at 0x7f85cf400000 EAL: PCI memory mapped at 0x7f85cf420000 PMD: eth_em_dev_init(): port_id 0 vendorID=0x8086 deviceID=0x15a2 I think this is not a NIC issue, but a general problem of the driver (or em code). [Wenzhuo] I don’t have a 15a2 on hand. I’m using i350. I haven’t hit the same problem. As you said it’s random. Would you like to let me know why you think it’s general not NIC specific? Thanks. About how to reproduce this problem, you mean use these CLIs, ethtool –s xxx advertise xxx, ethtool –s xxx duplex half/full, to change the peer port’s configuration? Correct. I modified l2fwd to check link status and print it on each port stats print iteration. Then from the peer I modified the link properties via ethtool. [Wenzhuo] Would you like to give me a patch of your change? Thanks. The result is that transitions from autoneg speeds and/or duplex mode settings are randomly not detected (rte_eth_link in rte_eth_dev_data is not updated), and it prints not-up-to-date state. Marc ^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [dpdk-dev] e1000: randomly loosing link change events triggered by the peer 2016-03-29 1:19 ` Lu, Wenzhuo @ 2016-03-30 8:04 ` Marc 0 siblings, 0 replies; 5+ messages in thread From: Marc @ 2016-03-30 8:04 UTC (permalink / raw) To: Lu, Wenzhuo; +Cc: dev On 29 March 2016 at 03:19, Lu, Wenzhuo <wenzhuo.lu@intel.com> wrote: > Hi Marc, > > > > *From:* marc.sune@gmail.com [mailto:marc.sune@gmail.com] *On Behalf Of * > Marc > *Sent:* Monday, March 28, 2016 7:03 PM > *To:* Lu, Wenzhuo > *Cc:* dev@dpdk.org > *Subject:* Re: e1000: randomly loosing link change events triggered by > the peer > > > > > > > > On 28 March 2016 at 03:54, Lu, Wenzhuo <wenzhuo.lu@intel.com> wrote: > > Hi Marc > > > > *From:* Marc Sune [mailto:marcdevel@gmail.com] > *Sent:* Saturday, March 26, 2016 9:43 AM > *To:* dev@dpdk.org; Lu, Wenzhuo > *Subject:* e1000: randomly loosing link change events triggered by the > peer > > > > I found that in the current HEAD in master testing it with an I218-LM in > autoneg mode, when link is forced to be renegociated by the peer (e.g. via > ethtool on a peer Linux box) _some_ change events are lost. > > > > It is quite random, but it seems to happen more while changing the speed > than when changing the duplex mode. > > > > However, when one or more link change events have been lost and the phy > medium is disconnected and reconnected, the link speed and duplex mode are > then correctly updated. > > > > Marc > > > > [Wenzhuo] Thanks for let us know this issue. May I ask some questions? Do > you mean this NIC 0x155A? > > > > 0x15A2, I218-LM (rev 03) > > > > EAL: PCI device 0000:00:19.0 on NUMA socket -1 > > EAL: probe driver: 8086:15a2 rte_em_pmd > > EAL: PCI memory mapped at 0x7f85cf400000 > > EAL: PCI memory mapped at 0x7f85cf420000 > > PMD: eth_em_dev_init(): port_id 0 vendorID=0x8086 deviceID=0x15a2 > > > > I think this is not a NIC issue, but a general problem of the driver (or > em code). > > [Wenzhuo] I don’t have a 15a2 on hand. I’m using i350. I haven’t hit the > same problem. As you said it’s random. Would you like to let me know why > you think it’s general not NIC specific? Thanks. > It was random but it was happening very frequently. > > > About how to reproduce this problem, you mean use these CLIs, ethtool –s > xxx advertise xxx, ethtool –s xxx duplex half/full, to change the peer > port’s configuration? > > > > Correct. I modified l2fwd to check link status and print it on each port > stats print iteration. Then from the peer I modified the link properties > via ethtool. > > [Wenzhuo] Would you like to give me a patch of your change? Thanks. > Unfortunately I checkedout that code, as it was a hack to test, but it was as simple as calling check_all_ports_link_status() in print_stats() for l2fwd example, so that was reporting link status periodically. Marc > > > The result is that transitions from autoneg speeds and/or duplex mode > settings are randomly not detected (rte_eth_link in rte_eth_dev_data is not > updated), and it prints not-up-to-date state. > > > > Marc > ^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2016-03-30 8:05 UTC | newest] Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- 2016-03-26 1:42 [dpdk-dev] e1000: randomly loosing link change events triggered by the peer Marc Sune 2016-03-28 1:54 ` Lu, Wenzhuo 2016-03-28 11:02 ` Marc 2016-03-29 1:19 ` Lu, Wenzhuo 2016-03-30 8:04 ` Marc
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for NNTP newsgroup(s).