Patch "e1000e: fix buffer overrun while the I219 is processing DMA transactions" has been added to the 4.14-stable tree
Subsystems (3)
This is a note to let you know that I've just added the patch titled
e1000e: fix buffer overrun while the I219 is processing DMA transactions
to the 4.14-stable tree which can be found at:
http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary
The filename of the patch is:
e1000e-fix-buffer-overrun-while-the-i219-is-processing-dma-transactions.patch
and it can be found in the queue-4.14 subdirectory.
If you, or anyone else, feels it should not be added to the stable tree, please let <[email protected]> know about it.
Quoted 1 line
From
b10effb92e272051dd1ec0d7be56bf9ca85ab927Mon Sep 17 00:00:00 2001
From: Sasha Neftin <[email protected]>
Date: Sun, 6 Aug 2017 16:49:18 +0300
Subject: e1000e: fix buffer overrun while the I219 is processing DMA transactions
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
From: Sasha Neftin <[email protected]>
commit b10effb92e272051dd1ec0d7be56bf9ca85ab927 upstream.
Intel® 100/200 Series Chipset platforms reduced the round-trip latency for the LAN Controller DMA accesses, causing in some high performance cases a buffer overrun while the I219 LAN Connected Device is processing the DMA transactions. I219LM and I219V devices can fall into unrecovered Tx hang under very stressfully UDP traffic and multiple reconnection of Ethernet cable. This Tx hang of the LAN Controller is only recovered if the system is rebooted. Slightly slow down DMA access by reducing the number of outstanding requests. This workaround could have an impact on TCP traffic performance on the platform. Disabling TSO eliminates performance loss for TCP traffic without a noticeable impact on CPU performance.
Please, refer to I218/I219 specification update:
https://www.intel.com/content/www/us/en/embedded/products/networking/
ethernet-connection-i218-family-documentation.html
--- drivers/net/ethernet/intel/e1000e/netdev.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-)
drivers/net/ethernet/intel/e1000e/netdev.c +5 -3
@@ -3021,8 +3021,8 @@ static void e1000_configure_tx(struct e1 hw->mac.ops.config_collision_dist (hw); -/* SPT and CNP Si errata workaround to avoid data corruption */ -if (hw->mac.type >= e1000_pch_spt) { +/* SPT and KBL Si errata workaround to avoid data corruption */ +if (hw->mac.type == e1000_pch_spt) {u32 reg_val; reg_val =er32 (IOSFPC );
@@ -3030,7 +3030,9 @@ static void e1000_configure_tx(struct e1ew32 (IOSFPC , reg_val); reg_val =er32 (TARC (0 )); - reg_val |=E1000_TARC0_CB_MULTIQ_3_REQ ; +/* SPT and KBL Si errata workaround to avoid Tx hang */ + reg_val &= ~BIT (28 ); + reg_val |=BIT (29 );ew32 (TARC (0 ), reg_val); } }
Patches currently in stable-queue which might be from [email protected] are
queue-4.14/e1000e-fix-buffer-overrun-while-the-i219-is-processing-dma-transactions.patch
Quoted 81 lines · On 28/11/2017 11:27, [email protected] wrote:
This is a note to let you know that I've just added the patch titled
e1000e: fix buffer overrun while the I219 is processing DMA transactions
to the 4.14-stable tree which can be found at:
http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summaryThe filename of the patch is:
e1000e-fix-buffer-overrun-while-the-i219-is-processing-dma-transactions.patch
and it can be found in the queue-4.14 subdirectory.If you, or anyone else, feels it should not be added to the stable tree, please let <[email protected]> know about it.
From
b10effb92e272051dd1ec0d7be56bf9ca85ab927Mon Sep 17 00:00:00 2001
From: Sasha Neftin <[email protected]>
Date: Sun, 6 Aug 2017 16:49:18 +0300
Subject: e1000e: fix buffer overrun while the I219 is processing DMA transactions
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bitFrom: Sasha Neftin <[email protected]>
commit
b10effb92e272051dd1ec0d7be56bf9ca85ab927upstream.Intel® 100/200 Series Chipset platforms reduced the round-trip latency for the LAN Controller DMA accesses, causing in some high performance cases a buffer overrun while the I219 LAN Connected Device is processing the DMA transactions. I219LM and I219V devices can fall into unrecovered Tx hang under very stressfully UDP traffic and multiple reconnection of Ethernet cable. This Tx hang of the LAN Controller is only recovered if the system is rebooted. Slightly slow down DMA access by reducing the number of outstanding requests. This workaround could have an impact on TCP traffic performance on the platform. Disabling TSO eliminates performance loss for TCP traffic without a noticeable impact on CPU performance.
Please, refer to I218/I219 specification update:
https://www.intel.com/content/www/us/en/embedded/products/networking/
ethernet-connection-i218-family-documentation.htmlSigned-off-by Sasha Neftin <[email protected]>Reviewed-by Dima Ruinskiy <[email protected]>Reviewed-by Raanan Avargil <[email protected]>Tested-by Aaron Brown <[email protected]>Signed-off-by Jeff Kirsher <[email protected]>Signed-off-by Amit Pundir <[email protected]>Signed-off-by Greg Kroah-Hartman <[email protected]>--- drivers/net/ethernet/intel/e1000e/netdev.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-)drivers/net/ethernet/intel/e1000e/netdev.c +5
-3@@ -3021,8 +3021,8 @@ static void e1000_configure_tx(struct e1 hw->mac.ops.config_collision_dist (hw); -/* SPT and CNP Si errata workaround to avoid data corruption */ -if (hw->mac.type >= e1000_pch_spt) { +/* SPT and KBL Si errata workaround to avoid data corruption */ +if (hw->mac.type == e1000_pch_spt) {u32 reg_val; reg_val =er32 (IOSFPC );@@ -3030,7 +3030,9 @@ static void e1000_configure_tx(struct e1ew32 (IOSFPC , reg_val); reg_val =er32 (TARC (0 )); - reg_val |=E1000_TARC0_CB_MULTIQ_3_REQ ; +/* SPT and KBL Si errata workaround to avoid Tx hang */ + reg_val &= ~BIT (28 ); + reg_val |=BIT (29 );ew32 (TARC (0 ), reg_val); } }
Quoted 3 lines
Patches currently in stable-queue which might be from [email protected] are
queue-4.14/e1000e-fix-buffer-overrun-while-the-i219-is-processing-dma-transactions.patch
This is good. To follow of this one we have sent (Jeff sent today) another one, same functionality but more fundamental and clear patch, please, refer to https://patchwork.ozlabs.org/patch/834472/. I believe you can take it too for stable build .
Sasha
Greg KH
maintainerQuoted 78 lines · Sasha
Quoted 77 lines · On 28/11/2017 11:27, [email protected] wrote:
This is a note to let you know that I've just added the patch titled
e1000e: fix buffer overrun while the I219 is processing DMA transactions
to the 4.14-stable tree which can be found at:
http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summaryThe filename of the patch is:
e1000e-fix-buffer-overrun-while-the-i219-is-processing-dma-transactions.patch
and it can be found in the queue-4.14 subdirectory.If you, or anyone else, feels it should not be added to the stable tree, please let <[email protected]> know about it.
From
b10effb92e272051dd1ec0d7be56bf9ca85ab927Mon Sep 17 00:00:00 2001
From: Sasha Neftin <[email protected]>
Date: Sun, 6 Aug 2017 16:49:18 +0300
Subject: e1000e: fix buffer overrun while the I219 is processing DMA transactions
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bitFrom: Sasha Neftin <[email protected]>
commit
b10effb92e272051dd1ec0d7be56bf9ca85ab927upstream.Intel� 100/200 Series Chipset platforms reduced the round-trip latency for the LAN Controller DMA accesses, causing in some high performance cases a buffer overrun while the I219 LAN Connected Device is processing the DMA transactions. I219LM and I219V devices can fall into unrecovered Tx hang under very stressfully UDP traffic and multiple reconnection of Ethernet cable. This Tx hang of the LAN Controller is only recovered if the system is rebooted. Slightly slow down DMA access by reducing the number of outstanding requests. This workaround could have an impact on TCP traffic performance on the platform. Disabling TSO eliminates performance loss for TCP traffic without a noticeable impact on CPU performance.
Please, refer to I218/I219 specification update:
https://www.intel.com/content/www/us/en/embedded/products/networking/
ethernet-connection-i218-family-documentation.htmlSigned-off-by Sasha Neftin <[email protected]>Reviewed-by Dima Ruinskiy <[email protected]>Reviewed-by Raanan Avargil <[email protected]>Tested-by Aaron Brown <[email protected]>Signed-off-by Jeff Kirsher <[email protected]>Signed-off-by Amit Pundir <[email protected]>Signed-off-by Greg Kroah-Hartman <[email protected]>--- drivers/net/ethernet/intel/e1000e/netdev.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-)drivers/net/ethernet/intel/e1000e/netdev.c +5
-3@@ -3021,8 +3021,8 @@ static void e1000_configure_tx(struct e1 hw->mac.ops.config_collision_dist (hw); -/* SPT and CNP Si errata workaround to avoid data corruption */ -if (hw->mac.type >= e1000_pch_spt) { +/* SPT and KBL Si errata workaround to avoid data corruption */ +if (hw->mac.type == e1000_pch_spt) {u32 reg_val; reg_val =er32 (IOSFPC );@@ -3030,7 +3030,9 @@ static void e1000_configure_tx(struct e1ew32 (IOSFPC , reg_val); reg_val =er32 (TARC (0 )); - reg_val |=E1000_TARC0_CB_MULTIQ_3_REQ ; +/* SPT and KBL Si errata workaround to avoid Tx hang */ + reg_val &= ~BIT (28 ); + reg_val |=BIT (29 );ew32 (TARC (0 ), reg_val); } }
Quoted 8 lines
Quoted 3 lines
Patches currently in stable-queue which might be from [email protected] are
queue-4.14/e1000e-fix-buffer-overrun-while-the-i219-is-processing-dma-transactions.patchThis is good. To follow of this one we have sent (Jeff sent today) another one, same functionality but more fundamental and clear patch, please, refer to https://patchwork.ozlabs.org/patch/834472/. I believe you can take it too for stable build .
What is the git commit id of the patch in Linus's tree?
You all do know how to properly tag patches to be included automatically in stable releases, right? If so, how come no one does this for this driver?
thanks,
greg k-h
Jeff Kirsher
Reviewed-by Alexander H Duyck Reviewed-by Dima Ruinskiy Reviewed-by Raanan Avargil Tested-by Aaron BrownQuoted 102 lines · Greg KH
Quoted 90 lines · Sasha
Quoted 89 lines · On 28/11/2017 11:27, [email protected] wrote:
This is a note to let you know that I've just added the patch titled
e1000e: fix buffer overrun while the I219 is processing DMA
transactionsto the 4.14-stable tree which can be found at:
http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queu
e.git;a=summaryThe filename of the patch is:
e1000e-fix-buffer-overrun-while-the-i219-is-processing-dma-transactions.patch
and it can be found in the queue-4.14 subdirectory.If you, or anyone else, feels it should not be added to the stable tree,
please let <[email protected]> know about it.From
b10effb92e272051dd1ec0d7be56bf9ca85ab927Mon Sep 17 00:00:00
2001
From: Sasha Neftin <[email protected]>
Date: Sun, 6 Aug 2017 16:49:18 +0300
Subject: e1000e: fix buffer overrun while the I219 is processing DMA transactions
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bitFrom: Sasha Neftin <[email protected]>
commit
b10effb92e272051dd1ec0d7be56bf9ca85ab927upstream.Intel® 100/200 Series Chipset platforms reduced the round-trip latency for the LAN Controller DMA accesses, causing in some high performance cases a buffer overrun while the I219 LAN Connected Device is processing the DMA transactions. I219LM and I219V devices can fall into unrecovered Tx hang under very stressfully UDP traffic and multiple reconnection of Ethernet cable. This Tx hang of the LAN Controller is only recovered if the system is rebooted. Slightly slow down DMA access by reducing the number of outstanding requests. This workaround could have an impact on TCP traffic performance on the platform. Disabling TSO eliminates performance loss for TCP traffic without a noticeable impact on CPU performance.
Please, refer to I218/I219 specification update:
https://www.intel.com/content/www/us/en/embedded/products/networking/
ethernet-connection-i218-family-documentation.htmlSigned-off-by Sasha Neftin <[email protected]>Reviewed-by Dima Ruinskiy <[email protected]>Reviewed-by Raanan Avargil <[email protected]>Tested-by Aaron Brown <[email protected]>Signed-off-by Jeff Kirsher <[email protected]>Signed-off-by Amit Pundir <[email protected]>Signed-off-by Greg Kroah-Hartman <[email protected]>--- drivers/net/ethernet/intel/e1000e/netdev.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-)drivers/net/ethernet/intel/e1000e/netdev.c +5
-3@@ -3021,8 +3021,8 @@ static void e1000_configure_tx(struct e1hw->mac.ops.config_collision_dist(hw); -/* SPT and CNP Si errata workaround to avoid data corruption */ -if (hw->mac.type >= e1000_pch_spt) { +/* SPT and KBL Si errata workaround to avoid data corruption */ +if (hw->mac.type == e1000_pch_spt) {u32 reg_val; reg_val =er32 (IOSFPC );@@ -3030,7 +3030,9 @@ static void e1000_configure_tx(struct e1ew32 (IOSFPC , reg_val); reg_val =er32 (TARC (0 )); -reg_val |= E1000_TARC0_CB_MULTIQ_3_REQ; +/* SPT and KBL Si errata workaround to avoid Tx hang */ + reg_val &= ~BIT (28 ); + reg_val |=BIT (29 );ew32 (TARC (0 ), reg_val); } }Patches currently in stable-queue which might be from sasha.neftin@in tel.com are
Quoted 10 lines
Quoted 2 lines
queue-4.14/e1000e-fix-buffer-overrun-while-the-i219-is-processing- dma-transactions.patch
This is good. To follow of this one we have sent (Jeff sent today) another
one, same functionality but more fundamental and clear patch, please, refer
to https://patchwork.ozlabs.org/patch/834472/. I believe you can take it too
for stable build .
Quoted 5 lines
What is the git commit id of the patch in Linus's tree?
You all do know how to properly tag patches to be included automatically in stable releases, right? If so, how come no one does this for this driver?
Sorry Greg, I did not realize the above commit was queued up for stable, which is why the follow on commit that Sasha was referring to was not CC'd
to stable mailing list. I do and will usual add the stable mailing list
to commits that need to be queued up for the stable trees.
Anyways, here is the commit id for the follow on patch for the changes above. David Miller just accepted it into his net tree this morning, here is the commit id info:
commit c0f4b163a03e73055dd734eaca64b9580e72e7fb
Author: Sasha Neftin <[email protected]> Date: Mon Nov 6 08:31:59 2017 +0200
e1000e: fix the use of magic numbers for buffer overrun issue
This is a follow on to commit b10effb92e27 ("fix buffer overrun while
the
I219 is processing DMA transactions") to address David Laights
concerns
about the use of "magic" numbers. So define masks as well as add additional code comments to give a better understanding of what needs
to
be done to avoid a buffer overrun.
Greg KH
maintainerQuoted 133 lines · Jeff Kirsher
Quoted 102 lines · Greg KH
Quoted 90 lines · Sasha
Quoted 89 lines · On 28/11/2017 11:27, [email protected] wrote:
This is a note to let you know that I've just added the patch titled
e1000e: fix buffer overrun while the I219 is processing DMA
transactionsto the 4.14-stable tree which can be found at:
http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queu
e.git;a=summaryThe filename of the patch is:
e1000e-fix-buffer-overrun-while-the-i219-is-processing-dma-transactions.patch
and it can be found in the queue-4.14 subdirectory.If you, or anyone else, feels it should not be added to the stable tree,
please let <[email protected]> know about it.From
b10effb92e272051dd1ec0d7be56bf9ca85ab927Mon Sep 17 00:00:00
2001
From: Sasha Neftin <[email protected]>
Date: Sun, 6 Aug 2017 16:49:18 +0300
Subject: e1000e: fix buffer overrun while the I219 is processing DMA transactions
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bitFrom: Sasha Neftin <[email protected]>
commit
b10effb92e272051dd1ec0d7be56bf9ca85ab927upstream.Intel� 100/200 Series Chipset platforms reduced the round-trip latency for the LAN Controller DMA accesses, causing in some high performance cases a buffer overrun while the I219 LAN Connected Device is processing the DMA transactions. I219LM and I219V devices can fall into unrecovered Tx hang under very stressfully UDP traffic and multiple reconnection of Ethernet cable. This Tx hang of the LAN Controller is only recovered if the system is rebooted. Slightly slow down DMA access by reducing the number of outstanding requests. This workaround could have an impact on TCP traffic performance on the platform. Disabling TSO eliminates performance loss for TCP traffic without a noticeable impact on CPU performance.
Please, refer to I218/I219 specification update:
https://www.intel.com/content/www/us/en/embedded/products/networking/
ethernet-connection-i218-family-documentation.htmlSigned-off-by Sasha Neftin <[email protected]>Reviewed-by Dima Ruinskiy <[email protected]>Reviewed-by Raanan Avargil <[email protected]>Tested-by Aaron Brown <[email protected]>Signed-off-by Jeff Kirsher <[email protected]>Signed-off-by Amit Pundir <[email protected]>Signed-off-by Greg Kroah-Hartman <[email protected]>--- drivers/net/ethernet/intel/e1000e/netdev.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-)drivers/net/ethernet/intel/e1000e/netdev.c +5
-3@@ -3021,8 +3021,8 @@ static void e1000_configure_tx(struct e1hw->mac.ops.config_collision_dist(hw); -/* SPT and CNP Si errata workaround to avoid data corruption */ -if (hw->mac.type >= e1000_pch_spt) { +/* SPT and KBL Si errata workaround to avoid data corruption */ +if (hw->mac.type == e1000_pch_spt) {u32 reg_val; reg_val =er32 (IOSFPC );@@ -3030,7 +3030,9 @@ static void e1000_configure_tx(struct e1ew32 (IOSFPC , reg_val); reg_val =er32 (TARC (0 )); -reg_val |= E1000_TARC0_CB_MULTIQ_3_REQ; +/* SPT and KBL Si errata workaround to avoid Tx hang */ + reg_val &= ~BIT (28 ); + reg_val |=BIT (29 );ew32 (TARC (0 ), reg_val); } }Patches currently in stable-queue which might be from sasha.neftin@in tel.com are
Quoted 10 lines
Quoted 2 lines
queue-4.14/e1000e-fix-buffer-overrun-while-the-i219-is-processing- dma-transactions.patch
This is good. To follow of this one we have sent (Jeff sent today) another
one, same functionality but more fundamental and clear patch, please, refer
to https://patchwork.ozlabs.org/patch/834472/. I believe you can take it too
for stable build .Quoted 5 lines
What is the git commit id of the patch in Linus's tree?
You all do know how to properly tag patches to be included automatically in stable releases, right? If so, how come no one does this for this driver?
Sorry Greg, I did not realize the above commit was queued up for stable, which is why the follow on commit that Sasha was referring to was not CC'd
to stable mailing list. I do and will usual add the stable mailing list
to commits that need to be queued up for the stable trees.Anyways, here is the commit id for the follow on patch for the changes above. David Miller just accepted it into his net tree this morning, here is the commit id info:
commit
c0f4b163a03e73055dd734eaca64b9580e72e7fbAuthor: Sasha Neftin <[email protected]> Date: Mon Nov 6 08:31:59 2017 +0200e1000e: fix the use of magic numbers for buffer overrun issue
This is a follow on to commit
b10effb92e27("fix buffer overrun while
the
I219 is processing DMA transactions") to address David Laights
concerns
about the use of "magic" numbers. So define masks as well as add additional code comments to give a better understanding of what needs
to
be done to avoid a buffer overrun.
Quoted 6 lines
Signed-off-by Sasha Neftin <[email protected]>Reviewed-by Alexander H Duyck <[email protected]>Reviewed-by Dima Ruinskiy <[email protected]>Reviewed-by Raanan Avargil <[email protected]>Tested-by Aaron Brown <[email protected]>Signed-off-by Jeff Kirsher <[email protected]>
Now added to the 4.14-stable tree, thanks.
greg k-h