linux-toradex.git/drivers/net/ethernet/chelsio, branch v3.11 Linux kernel for Apalis and Colibri modules Revert "cxgb3: Check and handle the dma mapping errors" 2013-08-15T08:23:32+00:00 Alexey Kardashevskiy aik@ozlabs.ru 2013-08-14T09:19:01+00:00 728e2ccaa3c4d20cf4d54b73a47956bf4d334a9f This reverts commit f83331bab149e29fa2c49cf102c0cd8c3f1ce9f9. As the tests PPC64 (powernv platform) show, IOMMU pages are leaking when transferring big amount of small packets (<=64 bytes), "ping -f" and waiting for 15 seconds is the simplest way to confirm the bug. Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Santosh Rastapur <santosh@chelsio.com> Cc: Jay Fenlason <fenlason@redhat.com> Cc: David S. Miller <davem@davemloft.net> Cc: Divy Le ray <divy@chelsio.com> Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru> Acked-by: Divy Le Ray <divy@chelsio.com> Signed-off-by: David S. Miller <davem@davemloft.net> 
This reverts commit f83331bab149e29fa2c49cf102c0cd8c3f1ce9f9.

As the tests PPC64 (powernv platform) show, IOMMU pages are leaking
when transferring big amount of small packets (<=64 bytes),
"ping -f" and waiting for 15 seconds is the simplest way to confirm the bug.

Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Santosh Rastapur <santosh@chelsio.com>
Cc: Jay Fenlason <fenlason@redhat.com>
Cc: David S. Miller <davem@davemloft.net>
Cc: Divy Le ray <divy@chelsio.com>
Signed-off-by: Alexey Kardashevskiy <aik@ozlabs.ru>
Acked-by: Divy Le Ray <divy@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
cxgb3: Missing rtnl lock in error recovery 2013-07-02T19:37:48+00:00 Benjamin Herrenschmidt benh@kernel.crashing.org 2013-06-30T04:37:11+00:00 7cc47d139f9a815a91bd9e7377063238c69a0423 When exercising error injection on IBM pseries machine, I hit the following warning: [ 251.450043] RTAS: event: 89, Type: Platform Error, Severity: 2 [ 253.549822] cxgb3 0006:01:00.0: enabling device (0140 -> 0142) [ 253.713560] cxgb3 0006:01:00.0: adapter recovering, PEX ERR 0x100 [ 254.895437] RTNL: assertion failed at net/core/dev.c (2031) [ 254.895467] CPU: 6 PID: 5449 Comm: eehd Tainted: G W 3.10.0-rc7-00157-gea461ab #19 [ 254.895474] Call Trace: [ 254.895483] [c000000fac56f7d0] [c000000000014dcc] .show_stack+0x7c/0x1f0 (unreliable) [ 254.895493] [c000000fac56f8a0] [c0000000007ba318] .dump_stack+0x28/0x3c [ 254.895500] [c000000fac56f910] [c0000000006c0384] .netif_set_real_num_tx_queues+0x224/0x230 [ 254.895515] [c000000fac56f9b0] [d00000000ef35510] .cxgb_open+0x80/0x3f0 [cxgb3] [ 254.895525] [c000000fac56fa50] [d00000000ef35914] .t3_resume_ports+0x94/0x100 [cxgb3] [ 254.895533] [c000000fac56fae0] [c00000000005fc8c] .eeh_report_resume+0x8c/0xd0 [ 254.895539] [c000000fac56fb60] [c00000000005e9fc] .eeh_pe_dev_traverse+0x9c/0x190 [ 254.895545] [c000000fac56fc10] [c000000000060000] .eeh_handle_event+0x110/0x330 [ 254.895551] [c000000fac56fca0] [c000000000060350] .eeh_event_handler+0x130/0x1a0 [ 254.895558] [c000000fac56fd30] [c0000000000ad758] .kthread+0xe8/0xf0 [ 254.895566] [c000000fac56fe30] [c00000000000a05c] .ret_from_kernel_thread+0x5c/0x80 It appears that t3_resume_ports() is called with the rtnl_lock held from the fatal error task but not from the PCI error callbacks. This fixes it. Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: David S. Miller <davem@davemloft.net> 
When exercising error injection on IBM pseries machine, I hit the
following warning:

[  251.450043] RTAS: event: 89, Type: Platform Error, Severity: 2
[  253.549822] cxgb3 0006:01:00.0: enabling device (0140 -> 0142)
[  253.713560] cxgb3 0006:01:00.0: adapter recovering, PEX ERR 0x100
[  254.895437] RTNL: assertion failed at net/core/dev.c (2031)
[  254.895467] CPU: 6 PID: 5449 Comm: eehd Tainted: G        W    3.10.0-rc7-00157-gea461ab #19
[  254.895474] Call Trace:
[  254.895483] [c000000fac56f7d0] [c000000000014dcc] .show_stack+0x7c/0x1f0 (unreliable)
[  254.895493] [c000000fac56f8a0] [c0000000007ba318] .dump_stack+0x28/0x3c
[  254.895500] [c000000fac56f910] [c0000000006c0384] .netif_set_real_num_tx_queues+0x224/0x230
[  254.895515] [c000000fac56f9b0] [d00000000ef35510] .cxgb_open+0x80/0x3f0 [cxgb3]
[  254.895525] [c000000fac56fa50] [d00000000ef35914] .t3_resume_ports+0x94/0x100 [cxgb3]
[  254.895533] [c000000fac56fae0] [c00000000005fc8c] .eeh_report_resume+0x8c/0xd0
[  254.895539] [c000000fac56fb60] [c00000000005e9fc] .eeh_pe_dev_traverse+0x9c/0x190
[  254.895545] [c000000fac56fc10] [c000000000060000] .eeh_handle_event+0x110/0x330
[  254.895551] [c000000fac56fca0] [c000000000060350] .eeh_event_handler+0x130/0x1a0
[  254.895558] [c000000fac56fd30] [c0000000000ad758] .kthread+0xe8/0xf0
[  254.895566] [c000000fac56fe30] [c00000000000a05c] .ret_from_kernel_thread+0x5c/0x80

It appears that t3_resume_ports() is called with the rtnl_lock held from
the fatal error task but not from the PCI error callbacks. This fixes it.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
cxgb4: Do not set net_device::dev_id to VI index 2013-06-13T08:24:17+00:00 Ben Hutchings bhutchings@solarflare.com 2013-06-10T16:34:13+00:00 8c367fcbe6549195d2eb11e62bea233f811aad41 net_device::dev_id should not be used merely to indicate a VI index, as it affects the way the local part of IPv6 addresses is normally generated. This field was intended for use where multiple devices may share a single assigned MAC address and need to have different IPv6 addresses. T4 VIs each have their own MAC address. Signed-off-by: Ben Hutchings <bhutchings@solarflare.com> Acked-by: Dimitris Michailidis <dm@chelsio.com> Signed-off-by: David S. Miller <davem@davemloft.net> 
net_device::dev_id should not be used merely to indicate a VI index,
as it affects the way the local part of IPv6 addresses is normally
generated.

This field was intended for use where multiple devices may share a
single assigned MAC address and need to have different IPv6 addresses.
T4 VIs each have their own MAC address.

Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Acked-by: Dimitris Michailidis <dm@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
cxgb3: Correct comparisons and calculations using skb->tail and skb-transport_header 2013-06-05T00:37:59+00:00 Li RongQing roy.qing.li@gmail.com 2013-06-03T22:11:16+00:00 15dd16c233b6d5ceea3a00f18d7c42902d0ac6b6 This corrects an regression introduced by "net: Use 16bits for *_headers fields of struct skbuff" when NET_SKBUFF_DATA_USES_OFFSET is not set. In that case skb->tail will be a pointer whereas skb->transport_header will be an offset from head. This is corrected by using wrappers that ensure that comparisons and calculations are always made using pointers. Cc: Simon Horman <horms@verge.net.au> Signed-off-by: Li RongQing <roy.qing.li@gmail.com> Reviewed-by: Simon Horman <horms@verge.net.au> Signed-off-by: David S. Miller <davem@davemloft.net> 
This corrects an regression introduced by "net: Use 16bits for *_headers
fields of struct skbuff" when NET_SKBUFF_DATA_USES_OFFSET is not set. In
that case skb->tail will be a pointer whereas skb->transport_header
will be an offset from head. This is corrected by using wrappers that
ensure that comparisons and calculations are always made using pointers.

Cc: Simon Horman <horms@verge.net.au>
Signed-off-by: Li RongQing <roy.qing.li@gmail.com>
Reviewed-by: Simon Horman <horms@verge.net.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
cxgb4: Force uninitialized state if FW_ON_ADAPTER is < FW_VERSION and we're the MASTER_PF 2013-06-01T00:25:43+00:00 Jay Hernandez jay@chelsio.com 2013-05-30T03:24:14+00:00 e69972f52bdc4bc384a21a6561d5673b4a94e989 Forcing uninitialized state allows us to upgrade and reinitialize the adapter. FW_VERSION_T4 = 1.4.0.0 FW_VERSION_T5 = 0.0.0.0 At this point driver supports above and greater than above version of firmware. If it doesn't find the required firmware version than it forces the adapter to be reinitialized as shown below. 1) If FW_ON_ADAPTER < FW_VERSION and we're the MASTER_PF force uninitialized state and a FW upgrade if available. - If FW_ON_ADAPTER < /lib/firmware/cxgb4/t*fw.bin we will update the adapters FW. - If FW_ON_ADAPTER >= /lib/firmware/cxgb4/t*fw.bin don't upgrade FW. - If upgrade_fw() fails force reinitialization of the adapter anyways, it might still work. Either way forcing the uninitialized state allows cxgb4 reinitialize FW. 2) If FW_ON_ADAPTER >= FW_VERSION driver follows normal path. Signed-off-by: Jay Hernandez <jay@chelsio.com> Signed-off-by: David S. Miller <davem@davemloft.net> 
Forcing uninitialized state allows us to upgrade and reinitialize the adapter.

FW_VERSION_T4 = 1.4.0.0
FW_VERSION_T5 = 0.0.0.0
At this point driver supports above and greater than above version of firmware.
If it doesn't find the required firmware version than it forces the adapter to
be reinitialized as shown below.

1) If FW_ON_ADAPTER < FW_VERSION and we're the MASTER_PF force uninitialized
   state and a FW upgrade if available.

       - If FW_ON_ADAPTER < /lib/firmware/cxgb4/t*fw.bin we will update the
         adapters FW.
       - If FW_ON_ADAPTER >= /lib/firmware/cxgb4/t*fw.bin don't upgrade FW.
       - If upgrade_fw() fails force reinitialization of the adapter anyways,
         it might still work.

   Either way forcing the uninitialized state allows cxgb4 reinitialize FW.

2) If FW_ON_ADAPTER >= FW_VERSION driver follows normal path.

Signed-off-by: Jay Hernandez <jay@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
cxgb3: Correct comparisons and calculations using skb->tail and skb-transport_header 2013-05-29T06:49:07+00:00 Simon Horman horms@verge.net.au 2013-05-28T20:34:24+00:00 be8b678c6c63010cd2b0f4f602bd3066f7827fb2 This corrects an regression introduced by "net: Use 16bits for *_headers fields of struct skbuff" when NET_SKBUFF_DATA_USES_OFFSET is not set. In that case skb->tail will be a pointer whereas skb->transport_header will be an offset from head. This is corrected by using wrappers that ensure that comparisons and calculations are always made using pointers. Signed-off-by: Simon Horman <horms@verge.net.au> Signed-off-by: David S. Miller <davem@davemloft.net> 
This corrects an regression introduced by "net: Use 16bits for *_headers
fields of struct skbuff" when NET_SKBUFF_DATA_USES_OFFSET is not set. In
that case skb->tail will be a pointer whereas skb->transport_header
will be an offset from head. This is corrected by using wrappers that
ensure that comparisons and calculations are always made using pointers.

Signed-off-by: Simon Horman <horms@verge.net.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
cxgb3: Check and handle the dma mapping errors 2013-05-23T07:03:54+00:00 Santosh Rastapur santosh@chelsio.com 2013-05-21T04:21:29+00:00 f83331bab149e29fa2c49cf102c0cd8c3f1ce9f9 This patch adds checks at approprate places whether *dma_map*() call has succeeded or not. Signed-off-by: Santosh Rastapur <santosh@chelsio.com> Reviewed-by: Jay Fenlason <fenlason@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net> 
This patch adds checks at approprate places whether *dma_map*() call has
succeeded or not.

Signed-off-by: Santosh Rastapur <santosh@chelsio.com>
Reviewed-by: Jay Fenlason <fenlason@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
cxgb3: Fix warning about using rcu_dereference when not in a rcu-locked section 2013-05-23T07:03:54+00:00 Jay Fenlason fenlason@redhat.com 2013-05-21T04:21:28+00:00 9313eb4be3220054c07a7c34c38764ea1ebcd220 It is about using rcu_dereference() when not in a rcu-locked section. It only happens on initialization hence fix the initialization to not rcu_dereference() Signed-off-by: Jay Fenlason <fenlason@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net> 
It is about using rcu_dereference() when not in a rcu-locked section. It only
happens on initialization hence fix the initialization to not rcu_dereference()

Signed-off-by: Jay Fenlason <fenlason@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
net/ethernet/chelsio/cxgb/cxgb2: Use module_pci_driver to register driver 2013-05-22T21:35:05+00:00 Peter Hüwe PeterHuewe@gmx.de 2013-05-21T13:42:55+00:00 65c21912b5b977f32a12336e87fe32f36105b981 Removing some boilerplate by using module_pci_driver instead of calling register and unregister in the otherwise empty init/exit functions. The name of the pci_driver struct had to be changed in order to prevent a build failure. Signed-off-by: Peter Huewe <peterhuewe@gmx.de> Signed-off-by: David S. Miller <davem@davemloft.net> 
Removing some boilerplate by using module_pci_driver instead of calling
register and unregister in the otherwise empty init/exit functions.

The name of the pci_driver struct had to be changed in order to prevent
a build failure.

Signed-off-by: Peter Huewe <peterhuewe@gmx.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
cxgb4: fix error recovery when t4_fw_hello returns a positive value 2013-05-03T20:10:34+00:00 Thadeu Lima de Souza Cascardo cascardo@linux.vnet.ibm.com 2013-05-03T08:11:04+00:00 777c2300865cb9b1b1791862ed23da677abfe6dc Since commit 636f9d371f70f22961fd598fe18380057518ca31 ("cxgb4: Add support for T4 configuration file"), t4_fw_hello may return a positive value instead of 0 for success. The recovery code tests only for zero and fails recovery for any other value. This fix tests for negative error values and fails only on those cases. Error recovery after an error injection works after this change. Signed-off-by: Thadeu Lima de Souza Cascardo <cascardo@linux.vnet.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net> 
Since commit 636f9d371f70f22961fd598fe18380057518ca31 ("cxgb4: Add
support for T4 configuration file"), t4_fw_hello may return a positive
value instead of 0 for success. The recovery code tests only for zero
and fails recovery for any other value.

This fix tests for negative error values and fails only on those cases.
Error recovery after an error injection works after this change.

Signed-off-by: Thadeu Lima de Souza Cascardo <cascardo@linux.vnet.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>