CC: stable@vger.stable.org Reported-by: Tobias Klausmann <klausman@schwarzvogel.de>
(cherry picked from commit b868179c47e9e8eadcd04c1f3105998e528988a3) Signed-off-by: Dean Nelson <dnelson@redhat.com> Tested-by: Tobias Klausmann <klausman@schwarzvogel.de> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Joe Jin <joe.jin@oracle.com>
Tushar Dave [Fri, 27 Jan 2012 09:00:46 +0000 (09:00 +0000)]
e1000: Adding e1000_dump function
When TX hang occurs e1000_dump prints TX ring, RX ring and Device registers.
(cherry picked from commit b04e36bac5075ceeacf11b639fbf0cb69aa68996) Signed-off-by: Tushar Dave <tushar.n.dave@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Joe Jin <joe.jin@oracle.com>
Dean Nelson [Fri, 16 Sep 2011 16:52:54 +0000 (16:52 +0000)]
e1000: don't enable dma receives until after dma address has been setup
Doing an 'ifconfig ethN down' followed by an 'ifconfig ethN up' on a qemu-kvm
guest system configured with two e1000 NICs can result in an 'unable to handle
kernel paging request at 0000000100000000' or 'bad page map in process ...' or
something similar.
These result from a 4096-byte page being corrupted with the following two-word
pattern (16-bytes) repeated throughout the entire page:
0x0000000000000000
0x0000000100000000
There can be other bits set as well. What is a constant is that the 2nd word
has the 32nd bit set. So one could see:
Which came from from a process' page table I dumped out when the marked line
was seen as bad by print_bad_pte().
The repeating pattern represents the e1000's two-word receive descriptor:
struct e1000_rx_desc {
__le64 buffer_addr; /* Address of the descriptor's data buffer */
__le16 length; /* Length of data DMAed into data buffer */
__le16 csum; /* Packet checksum */
u8 status; /* Descriptor status */
u8 errors; /* Descriptor Errors */
__le16 special;
};
And the 32nd bit of the 2nd word maps to the 'u8 status' member, and
corresponds to E1000_RXD_STAT_DD which indicates the descriptor is done.
The corruption appears to result from the following...
. An 'ifconfig ethN down' gets us into e1000_close(), which through a number
of subfunctions results in:
1. E1000_RCTL_EN being cleared in RCTL register. [e1000_down()]
2. dma_free_coherent() being called. [e1000_free_rx_resources()]
. An 'ifconfig ethN up' gets us into e1000_open(), which through a number of
subfunctions results in:
1. dma_alloc_coherent() being called. [e1000_setup_rx_resources()]
2. E1000_RCTL_EN being set in RCTL register. [e1000_setup_rctl()]
3. E1000_RCTL_EN being cleared in RCTL register. [e1000_configure_rx()]
4. RDLEN, RDBAH and RDBAL registers being set to reflect the dma page
allocated in step 1. [e1000_configure_rx()]
5. E1000_RCTL_EN being set in RCTL register. [e1000_configure_rx()]
During the 'ifconfig ethN up' there is a window opened, starting in step 2
where the receives are enabled up until they are disabled in step 3, in which
the address of the receive descriptor dma page known by the NIC is still the
previous one which was freed during the 'ifconfig ethN down'. If this memory
has been reallocated for some other use and the NIC feels so inclined, it will
write to that former dma page with predictably unpleasant results.
I realize that in the guest, we're dealing with an e1000 NIC that is software
emulated by qemu-kvm. The problem doesn't appear to occur on bare-metal. Andy
suspects that this is because in the emulator link-up is essentially instant
and traffic can start flowing immediately. Whereas on bare-metal, link-up
usually seems to take at least a few milliseconds. And this might be enough
to prevent traffic from flowing into the device inside the window where
E1000_RCTL_EN is set.
So perhaps a modification needs to be made to the qemu-kvm e1000 NIC emulator
to delay the link-up. But in defense of the emulator, it seems like a bad idea
to enable dma operations before the address of the memory to be involved has
been made known.
The following patch no longer enables receives in e1000_setup_rctl() but leaves
them however they were. It only enables receives in e1000_configure_rx(), and
only after the dma address has been made known to the hardware.
There are two places where e1000_setup_rctl() gets called. The one in
e1000_configure() is followed immediately by a call to e1000_configure_rx(), so
there's really no change functionally (except for the removal of the problem
window. The other is in __e1000_shutdown() and is not followed by a call to
e1000_configure_rx(), so there is a change functionally. But consider...
. An 'ifconfig ethN down' (just as described above).
. A 'suspend' of the system, which (I'm assuming) will find its way into
e1000_suspend() which calls __e1000_shutdown() resulting in:
1. E1000_RCTL_EN being set in RCTL register. [e1000_setup_rctl()]
And again we've re-opened the problem window for some unknown amount of time.
(cherry picked from commit d5bc77a223b0e9b9dfb002048d2b34a79e7d0b48) Signed-off-by: Andy Gospodarek <andy@greyhouse.net> Signed-off-by: Dean Nelson <dnelson@redhat.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Joe Jin <joe.jin@oracle.com>
Dean Nelson [Thu, 25 Aug 2011 14:39:24 +0000 (14:39 +0000)]
e1000: save skb counts in TX to avoid cache misses
Virtual Machines with emulated e1000 network adapter running on Parallels'
server were seeing kernel panics due to the e1000 driver dereferencing an
unexpected NULL pointer retrieved from buffer_info->skb.
The problem has been addressed for the e1000e driver, but not for the e1000.
Since the two drivers share similar code in the affected area, a port of the
following e1000e driver commit solves the issue for the e1000 driver:
e1000e: save skb counts in TX to avoid cache misses
In e1000_tx_map, precompute number of segements and bytecounts which
are derived from fields in skb; these are stored in buffer_info. When
cleaning tx in e1000_clean_tx_irq use the values in the associated
buffer_info for statistics counting, this eliminates cache misses
on skb fields.
(cherry picked from commit 31c15a2f24ebdab14333d9bf5df49757842ae2ec) Signed-off-by: Dean Nelson <dnelson@redhat.com> Acked-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Joe Jin <joe.jin@oracle.com>
Florian Fainelli [Wed, 4 Jan 2012 20:23:35 +0000 (20:23 +0000)]
e1000: cleanup CE4100 MDIO registers access
A global variable is currently used to hold the virtual address of the
CE4100 MDIO base register address. Store the address in the e1000_hw
structure and update macros accordingly.
(cherry picked from commit 13acde8fffc0afbe8341fe08d2c594243f905c1f) Signed-off-by: Florian Fainelli <ffainelli@freebox.fr> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Joe Jin <joe.jin@oracle.com>
Florian Fainelli [Wed, 4 Jan 2012 20:23:34 +0000 (20:23 +0000)]
e1000: unmap ce4100_gbe_mdio_base_virt in e1000_remove
We are not unmapping ce4100_gbe_mdio_base_virt in exit path in case
we are running on a CE4100 adapter, fix that.
(cherry picked from commit 1c26750c485de20a3bd68ec5d396887650b53976) Signed-off-by: Florian Fainelli <ffainelli@freebox.fr> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Joe Jin <joe.jin@oracle.com>
Jesse Brandeburg [Wed, 4 Jan 2012 20:23:33 +0000 (20:23 +0000)]
e1000: fix lockdep splat in shutdown handler
As reported by Steven Rostedt, e1000 has a lockdep splat added
during the recent merge window. The issue is that
cancel_delayed_work is called while holding our private mutex.
There is no reason that I can see to hold the mutex during pci
shutdown, it was more just paranoia that I put the mutex_lock
around the call to e1000_down.
In a quick survey lots of drivers handle locking differently when
being called by the pci layer. The assumption here is that we
don't need the mutexes' protection in this function because
the driver could not be unloaded while in the shutdown handler
which is only called at reboot or poweroff.
Reported-by: Steven Rostedt <rostedt@goodmis.org>
(cherry picked from commit 3a3847e007aae732d64d8fd1374126393e9879a3) Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> Tested-by: Steven Rostedt <rostedt@goodmis.org> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Joe Jin <joe.jin@oracle.com>
I merely installed coccinelle, read the documentation and took credit.
(backported from commit 3db1cd5c05f35fb43eb134df6f321de4e63141f2) Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Joe Jin <joe.jin@oracle.com>
Joe Perches [Tue, 20 Sep 2011 15:32:52 +0000 (15:32 +0000)]
intel: Convert <FOO>_LENGTH_OF_ADDRESS to ETH_ALEN
Use the normal #defines not module specific ones.
(cherry picked from commit ea99d832cce7e724ba37c488e0571a00cb14d430) Signed-off-by: Joe Perches <joe@perches.com> Tested-by: Stephen Ko <stephen.s.ko@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Joe Jin <joe.jin@oracle.com>
Eric Dumazet [Thu, 13 Oct 2011 07:53:42 +0000 (07:53 +0000)]
e1000: fix skb truesize underestimation
e1000 allocates a full page per skb fragment. We must account PAGE_SIZE
increments on skb->truesize, not the actual frag length.
(cherry picked from commit ed64b3cc11502f50e1401f12e33d021592800bca) Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Joe Jin <joe.jin@oracle.com>
Jesse Brandeburg [Wed, 5 Oct 2011 07:24:51 +0000 (07:24 +0000)]
e1000: convert to private mutex from rtnl
The e1000 driver when running with lockdep could run into
some possible deadlocks between the work items acquiring
rtnl and the rtnl lock being acquired before work items
were cancelled.
Use a private mutex to make sure lock ordering isn't violated.
The private mutex is only used to protect areas not generally
covered by the rtnl lock already.
(cherry picked from commit 0ef4eedc2e98edd51cd106e1f6a27178622b7e57) Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> CC: Thomas Gleixner <tglx@linutronix.de> CC: Tushar Dave <tushar.n.dave@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Joe Jin <joe.jin@oracle.com>
Jesse Brandeburg [Wed, 5 Oct 2011 07:24:46 +0000 (07:24 +0000)]
e1000: convert mdelay to msleep
With the previous commit, there are several functions
that are only ever called from thread context, and are
able to sleep with msleep instead of mdelay.
(cherry picked from commit 4e0d8f7d97f9150bdd07f6355e5c1486967dce79) Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> CC: Thomas Gleixner <tglx@linutronix.de> CC: Tushar Dave <tushar.n.dave@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Joe Jin <joe.jin@oracle.com>
Jesse Brandeburg [Wed, 5 Oct 2011 07:24:41 +0000 (07:24 +0000)]
e1000: convert hardware management from timers to threads
Thomas Gleixner (tglx) reported that e1000 was delaying for many milliseconds
(using mdelay) from inside timer/interrupt context. None of these paths are
performance critical and can be moved into threads/work items. This patch
implements the work items and the next patch changes the mdelays to msleeps.
(cherry picked from commit a4010afef585b7142eb605e3a6e4210c0e1b2957) Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> CC: Thomas Gleixner <tglx@linutronix.de> CC: Tushar Dave <tushar.n.dave@intel.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Joe Jin <joe.jin@oracle.com>
Joe Jin [Thu, 17 May 2012 03:02:23 +0000 (11:02 +0800)]
e100: Remove alloc_etherdev error messages
alloc_etherdev has a generic OOM/unable to alloc message.
Remove the duplicative messages after alloc_etherdev calls.
(backported from commit 41de8d4cff21a2e81e3d9ff66f5f7c903f9c3ab1) Signed-off-by: Joe Perches <joe@perches.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Joe Jin <joe.jin@oracle.com>
Joe Jin [Thu, 17 May 2012 02:52:30 +0000 (10:52 +0800)]
net: Remove Intel NICs unnecessary driver assignments of ethtool_ringparam fields to zero
For e100/e1000/e1000e/igb/igbvf/ixgb/ixgbe/ixgbevf
Per comments from Ben Hutchings on a previous patch, sweep the floors
a little removing unnecessary assignments of zero to fields of struct
ethtool_ringparam in driver code supporting ethtool -g.
(backported from commit 8b0c11679fd37522d8d34a76101319a085d80912) Signed-off-by: Rick Jones <rick.jones2@hp.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Joe Jin <joe.jin@oracle.com>
Ben Greear [Fri, 17 Feb 2012 13:44:28 +0000 (13:44 +0000)]
e100: Show short v/s long rx length errors in ethtool stats.
(cherry picked from commit 6f66342c1ecd59467d4579176dd81a4e837a06ef) Signed-off-by: Ben Greear <greearb@candelatech.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Joe Jin <joe.jin@oracle.com>
Ben Greear [Fri, 17 Feb 2012 13:44:23 +0000 (13:44 +0000)]
e100: Fix rx-over-length statistics.
The old code would += the total errors every time
stats were gathered. Instead, keep a count of short-pkt
and long-pkt counters and then simply add them together
for the rx-over-length stat.
(cherry picked from commit d24d65eda97fe51f2996538148e85d309e2460e4) Signed-off-by: Ben Greear <greearb@candelatech.com> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Joe Jin <joe.jin@oracle.com>
Jesse Brandeburg [Sat, 22 Oct 2011 05:18:10 +0000 (05:18 +0000)]
e100: make sure vlan support isn't advertised on old adapters
e100 parts don't support vlan offload but they generally do
allow use of vlans in higher software layers via the 8021q module.
That said, there are a couple of really old revisions of e100
hardware that don't even allow the longer frame sizes
required for vlan use with standard MTU.
Use the VLAN_CHALLENGED flag to prevent vlan binding to these
devices.
Reported-by: Michael Tokarev <mjt@tls.msk.ru>
(cherry picked from commit 243559f436f26b571ea3a4e70ff082892dc58f16) Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com> CC: Michael Tokarev <mjt@tls.msk.ru> CC: David Lamparter <equinox@diac24.net> Tested-by: Aaron Brown <aaron.f.brown@intel.com> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com> Signed-off-by: Joe Jin <joe.jin@oracle.com>
Jason Wang [Fri, 30 Dec 2011 23:44:42 +0000 (23:44 +0000)]
8139cp: properly config rx mode after resuming
Rx mode should be reset after resming, so unconditionally updating rx
mode rather than conditionally updating based on the value we
remembered, otherwise unexpected value may be used by the nic after
resuming.
btw. I find and test this when debugging guest hibernation in qemu, as
I did not have a 8139cp card in hand, this patch is untested in a
physical 8139cp card, plase review it carefully.
(cherry picked from commit f872b237c1750221932e715da2552225afe4a95c) Signed-off-by: Jason Wang <jasowang@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Joe Jin <joe.jin@oracle.com>
Neil Horman [Fri, 9 Mar 2012 22:50:13 +0000 (14:50 -0800)]
bnx2fc: Remove bh disable in softirq context
As with the fcoe sw transport, the bnx2fc packet handler function runs only in
softirq context. Theres no need to disable bottom halves here
Signed-off-by: Neil Horman <nhorman@tuxdriver.com> Acked-by: Bhanu Prakash Gollapudi <bprakash@broadcom.com> Signed-off-by: Robert Love <robert.w.love@intel.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
(cherry picked from commit fc05ab74b78a9e16b0faa9f0dc3c87f3f30d3231)
Signed-off-by: Bhanu Prakash Gollapudi <bprakash@broadcom.com> Signed-off-by: Dmitry Kravkov <dmitry@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
(cherry picked from commit 1101a0d87a8e336914d2f31b528748a231c375ed)
bnx2fc: NPIV ports go offline when interface is brought down & up
When there are 255 NPIV ports, and the interface is brought down & up, both
physical and NPIV ports are logged off and never logged back in. Since
discovery happens on single CPU, XID resources on that CPU will be limited,
which when exhausted the discovery fails. Increase the XID resource range to
ensure that the discovery completes successfully. Also ensure that
fc_exch_mgr_alloc() doesn't fail on the system that has lower number of CPUs.
Signed-off-by: Bhanu Prakash Gollapudi <bprakash@broadcom.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
(cherry picked from commit 7d742f659e00f08016a4446a9134036e47f9a0cf)
Host drops sessions when flood of unsolicited LOGOs are received from the
target. Because of unsufficient PLOGI retries, upon exceeding the retry count
of 3, the target sessions are dropped. Increased the retry count to 255 to
allow sufficient retries in this scenario.
Signed-off-by: Bhanu Prakash Gollapudi <bprakash@broadcom.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
(cherry picked from commit 44c570b5b09d1d1cc3167834f89f754f1bc5ac14)
System panics while accessing stale pointer - timer_work_queue - in the IO path
before bnx2fc_stop is called. Fix is to destroy the workqueue after the destroy
operation is complete.
Signed-off-by: Bhanu Prakash Gollapudi <bprakash@broadcom.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
(cherry picked from commit 2a7b29c5f267c143e280c1a5174a0d930bd6c1e3)
When SRR LS_ACC is dropped, the driver was not issuing ABTS for SRR when it
times out. Since the target received SRR, it was able to send the XFER_RDY and
the the original IO request completed successfully. In this condition ABTS was
not sent during bnx2fc_srr_compl(). Fix this by first checking for ELS timeout
and issue ABTS before checking if original IO request is complete.
Signed-off-by: Bhanu Prakash Gollapudi <bprakash@broadcom.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
(cherry picked from commit 32c30454507b4f5f00661ac12ddbcc150983b9ff)
If the IO and the corresponding ABTS are not responded by a target, cleanup the
IO and issue explicit logout when ulp timer expires while waiting for ABTS to
complete. Wait for the session to be ready before returning to the SCSI layer.
If the session is not ready let the SCSI-ml escalate the error recovery.
Signed-off-by: Bhanu Prakash Gollapudi <bprakash@broadcom.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
(cherry picked from commit 99cc600cdd6f938633394523447378f7a43f4340)
bnx2fc: call ctlr_link_up only when the interface is enabled
Link may not be up when the driver receives ulp_start event, and hence
fcoe_ctlr_link_up is not called. Call fcoe_ctlr_link_up during
indicate_netevent only when the interface is enabled. (It has to be called when
enabled because that is an indication that the vlan discovery is completed).
Signed-off-by: Bhanu Prakash Gollapudi <bprakash@broadcom.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
(cherry picked from commit fd8fa9071e49a56cc91f739813ea88f16b7c1240)
Firmware asserts when the same CQE is armed twice. This scenario happens during
RSCN stress tests as driver incorrects arms the CQ after the session is
offloaded.
Signed-off-by: Bhanu Prakash Gollapudi <bprakash@broadcom.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
(cherry picked from commit a96e8e11635773dd88dbe1307f9482146cfe0562)
bnx2fc: Fix panic caused because of incorrect errror handling in create().
Driver incorrectly calls bnx2fc_interface_cleanup() when bnx2fc_if_create fails
which accesses bad pointer. Handle bnx2fc_if_create failure by directly calling
bnx2fc_net_cleanup.
Signed-off-by: Bhanu Prakash Gollapudi <bprakash@broadcom.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
(cherry picked from commit 013068fa6f65f7037adedc141fbd27afc99ab1bb)
bnx2fc: Avoid calling bnx2fc_if_destroy with unnecessary locks
It is not required to hold rtnl_lock and bnx2fc_dev_lock when calling
bnx2fc_if_destroy, as the locking is only required to serialize creation and
deletion of fcoe instances. More importantly, this unnecessary locking causes
deadlock as bnx2fc_if_destroy calls fc_remove_host holding rtnl_lock.
Signed-off-by: Bhanu Prakash Gollapudi <bprakash@broadcom.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
(cherry picked from commit 0cbf32e1681d870632a1772601cbaadd996dc978)
bnx2fc: Validate vlan id in NETDEV_UNREGISTER handler
When bnx2fc receives an UNREGISTER event on a vlan interface it calls
destroy on all interfaces that matches the physical interface. Add
vlan_id check to destroy only the vlan interface that generated the
event.
Signed-off-by: Nithin Nayak Sujir <nsujir@broadcom.com> Signed-off-by: Bhanu Prakash Gollapudi <bprakash@broadcom.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
(cherry picked from commit 26b2982f78c1fc6f486a67271b1d0a0d305dd54b)
ABTS was not issued for timed out REC, as REC completion handler exits out if
the IO completed. Check for timed out REC and issue ABTS before proceeding with
further processing in REC completion handler. Also, initialize rec_retry and
srr_retry before starting the IO.
Signed-off-by: Bhanu Prakash Gollapudi <bprakash@broadcom.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
(cherry picked from commit c1c16bd51a29eea8843f20161ddd32cddc524142)
bnx2fc: Send solicitation only after vlan discovery is complete
Link up event is generated to the driver even before vlan discovery has
started. Because of this driver can send discovery solicitation on a stale
vlan. Call fcoe_ctlr_link_up() only when the driver is in enabled state, which
implies the vlan discovery is complete before sending solicitation.
Signed-off-by: Bhanu Prakash Gollapudi <bprakash@broadcom.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
(cherry picked from commit 8a5badf1ea10c726b9cc04e52f91395b1248e034)
bnx2fc: Prevent creating of NPIV port with duplicate WWN
This patch adds a validation step before allowing creation of a new NPIV port.
It checks whether the WWPN passed for the new NPIV port to be created is unique
for the given physical port.
Signed-off-by: Bhanu Prakash Gollapudi <bprakash@broadcom.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
(cherry picked from commit 861efc547ce517e3d75a507ebc2268e6a0dfa767)
[SCSI] fcoe,libfcoe: Move common code for fcoe_get_lesb to fcoe_transport
Except for obtaining the netdev from lport, fcoe_get_lesb is the common code
for the LLDs.
Signed-off-by: Bhanu Prakash Gollapudi <bprakash@broadcom.com> Acked-by: Yi Zou <yi.zou@intel.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
(cherry picked from commit 814740d5f67ae5f205349019bfaae38bcd0c8732)
[SCSI] bnx2fc: code cleanup in bnx2fc_offload_session
- Free session resources before rport logoff
- Do not free session resources in bnx2fc_alloc_session_resc() as it is handled
in caller's error handling path.
- Do not call bnx2fc_free_session_resc() if bnx2fc_init_tgt() fails as cq_lock
is not yet initialized.
Signed-off-by: Bhanu Prakash Gollapudi <bprakash@broadcom.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
(cherry picked from commit 5fb8fd0dbc0f7e3bc3ea594223e5d4b50c2092d9)
[SCSI] bnx2fc: Fix NULL pointer deref during arm_cq.
There exists a race condition between CQ doorbell unmap and IO completion path
that arms the CQ which causes a NULL dereference. Protect the ctx_base with
cq_lock to avoid this. Also, wait for the CQ doorbell to be successfully mapped
before arming the CQ.
Also, do not count uncolicited CQ completions for free_sqes.
Signed-off-by: Bhanu Prakash Gollapudi <bprakash@broadcom.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
(cherry picked from commit b338c785c5c945383046ff39092e3021ea5b1d95)
[SCSI] bnx2fc: IO errors when receiving unsolicited LOGO
During the unsolicited LOGO processing, the session is uploaded and offloaded
after the relogin is complete. In between any new IOs are errored back as the
upload completion flag is set. Upon exhausting the retry count, the application
fails the IOs. Return target busy for all the cases when session is not ready.
Signed-off-by: Bhanu Prakash Gollapudi <bprakash@broadcom.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
(cherry picked from commit 81214013130cd24142f6465f7f5a256fed530c17)
[SCSI] bnx2fc: Do not reuse the fcoe connection id immediately
CFC_DELETE is issued 2 secs after CONN_TERM is completed. If the session is
uploaded and offloaded immediately, it has to wait for the connection id to be
available.
Signed-off-by: Bhanu Prakash Gollapudi <bprakash@broadcom.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
(cherry picked from commit 291fbe138f820f880b3e70852d200c0f2527430c)
[SCSI] bnx2fc: Handle NETDEV_UNREGISTER for vlan devices
Since the driver holds the reference for vlan netdev, the reference has to be
released by the driver when the vlan device is removed. Driver handles this in
NETDEV_UNREGISTER event.
Signed-off-by: Bhanu Prakash Gollapudi <bprakash@broadcom.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
(cherry picked from commit abc49a937b23adc7d9748709ca82ee32464ae089)
[SCSI] bnx2fc: Reorganize cleanup code between interface_cleanup and if_destory
Move interface specific cleanup functionality to from bnx2fc_if_destroy to
bnx2fc_interface_cleanup. Do not access interface/hba in bnx2fc_if_destroy as
by the time this function is called interface may already be destroyed. This
patch is in preparation to handle NETDEV_UNREGISTER on a vlan device.
Signed-off-by: Bhanu Prakash Gollapudi <bprakash@broadcom.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
(cherry picked from commit 9be17fc43e0121e966049a323ad38a35626525c1)
[SCSI] bnx2fc: Change function names of bnx2fc_netdev_setup/bnx2fc_netdev_cleanup
Change them to bnx2fc_interface_setup/bnx2fc_interface_cleanup in preperation
for the patches to follow. Interface specific cleanup functionality will be
moved to bnx2fc_interface_cleanup.
Signed-off-by: Bhanu Prakash Gollapudi <bprakash@broadcom.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
(cherry picked from commit 776cebcac6fad2f638c0ab16e2111a84c1c85d84)
[SCSI] bnx2fc: Do not attempt destroying NPIV port twice
When NPIV ports are created/deleted rapidly there is a race condition between
bnx2fc_vport_destroy() from sysfs and bnx2fc_flogi_resp(), which could try to
delete the NPIV port from the list twice. Fix is to loop through the list of
NPIV ports to find a match, and only when it exists remove it.
Signed-off-by: Bhanu Prakash Gollapudi <bprakash@broadcom.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
(cherry picked from commit cdf54668bc48193e43adc8f75ce419ce0ce50fc0)
Eric Dumazet [Thu, 2 Feb 2012 13:03:22 +0000 (14:03 +0100)]
[SCSI] bnx2i: use kthread_create_on_node()
bnx2i_percpu_thread_create() create per cpu kthread, and should use
proper NUMA aware API.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Acked-by: Eddie Wai <eddie.wai@broadcom.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
(cherry picked from commit 93713f3bf1d74c57ff851d17de54bb273d94210c)
Eddie Wai [Wed, 7 Dec 2011 06:41:21 +0000 (22:41 -0800)]
[SCSI] bnx2i: Fixed kernel panic caused by unprotected task->sc->request deref
During session recovery, the conn_stop call will trigger a flush
to all outstanding SCSI cmds in the xmit queue. This will set
all outstanding task->sc to NULL prior to the session_teardown
call which frees the task memory.
In the bnx2i SCSI response processing path, only the task was being checked
for NULL under the session lock before the task->sc->request dereferencing.
If there are outstanding SCSI cmd responses pending for process, the
following kernel panic can be exposed where task->sc was found to be NULL.
This patch adds the !task->sc check and also protects the sc dereferencing
under the session lock.
Signed-off-by: Eddie Wai <eddie.wai@broadcom.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
(cherry picked from commit a878185c3b93e692ace0d1628a47f3d75504ab4f)
Eddie Wai [Fri, 26 Aug 2011 18:16:47 +0000 (11:16 -0700)]
[SCSI] bnx2i: Fixed the endian on TTT for NOP out transmission
The iscsi_nopout task's TTT is defined as __be32 while the DMA
memory to the chip is CPU specific. This creates a problem for
unsolicited NOP-In responses where the TTT is not the RESERVED
tag of 0xFFs. This patch adds a call to be32_to_cpu for the TTT
specified.
Signed-off-by: Eddie Wai <eddie.wai@broadcom.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
(cherry picked from commit 610602f369b4c810c9df05e431abd38f38cb8e0d)
The problem lies in the slow path sg_io (and perhaps sg_scsi_ioctl) call to
blk_get_request->get_request/wait->blk_alloc_request->blk_rq_init which
re-initializes the request->cpu to -1. There is no assignment for cpu from
that to the request_fn call to low level drivers.
When this happens, the sc->request->cpu will be using the init value of
-1. This will create a kernel panic when it hits bnx2i because the code
refers it to get the per_cpu variables ptr.
This change is to put in a guard against that and also for cases when
bio affinity/queue completion to the same cpu is not enabled. In those
cases, the request->cpu will remain a -1 also.
For the case when the blk layer did not setup the request->cpu, bnx2i
will complete the sc with the current CPU of the thread.
Signed-off-by: Eddie Wai <eddie.wai@broadcom.com> Signed-off-by: James Bottomley <JBottomley@Parallels.com>
(cherry picked from commit 0d83ab65ff1b54ce8b6cd172285cf71a38c4cceb)
David S. Miller [Fri, 12 Aug 2011 10:00:47 +0000 (03:00 -0700)]
cnic: Fix select dependencies in bnx2fc/bnx2i Kconfig.
Reported-by: Stephen Rothwell <sfr@canb.auug.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
(cherry picked from commit 4c78893b3d107e2a053c8f51c526510857c09858)
Signed-off-by: Yaniv Rosner <yanivr@broadcom.com> Signed-off-by: Eilon Greenstein <eilong@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
(cherry picked from commit ab505dec96340946079d1288f49041bea9f259ff)
Enable FEC(Forward Error Correction) for 57810-KR to reduce link errors.
Signed-off-by: Yaniv Rosner <yanivr@broadcom.com> Signed-off-by: Eilon Greenstein <eilong@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
(cherry picked from commit 6b1f3900fc0909fbf3bd672242378015f76b3df8)
Dmitry Kravkov [Tue, 9 Aug 2011 03:11:13 +0000 (03:11 +0000)]
bnx2x: disable dcb on 578xx since not supported yet
Signed-off-by: Dmitry Kravkov <dmitry@broadcom.com> Signed-off-by: Eilon Greenstein <eilong@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
(cherry picked from commit 7712b644964c36d1216315b791faccc7d5238b16)
Dmitry Kravkov [Tue, 30 Aug 2011 00:08:40 +0000 (00:08 +0000)]
bnx2x: decrease print level to debug
It may happen every link toggle.
Signed-off-by: Dmitry Kravkov <dmitry@broadcom.com> Signed-off-by: Eilon Greenstein <eilong@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
(cherry picked from commit 5f837363457a2280530373267f86092625d15a4d)
Dmitry Kravkov [Tue, 30 Aug 2011 00:08:39 +0000 (00:08 +0000)]
bnx2x: fix BRB thresholds for dropless_fc mode
Fix the thresholds according to 5778x HW and increase rx_ring size
to suit new thresholds in dropless_fc mode.
Signed-off-by: Dmitry Kravkov <dmitry@broadcom.com> Signed-off-by: Eilon Greenstein <eilong@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
(cherry picked from commit dfacf1387ceb6d7d6df614b18016fd1f347a1996)
Dmitry Kravkov [Sun, 9 Oct 2011 23:57:36 +0000 (23:57 +0000)]
bnx2x: fix cl_id allocation for non-eth clients for NPAR mode
There are some consolidations of NPAR configuration
when FCoE and iSCSI L2 clients will get the same id,
in this case FCoE ring will be non-functional.
Signed-off-by: Dmitry Kravkov <dmitry@broadcom.com> Signed-off-by: Eilon Greenstein <eilong@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
(cherry picked from commit 134d0f974c55c942541f1663d51fa846257eff2a)
Vladislav Zolotarov [Tue, 9 Aug 2011 03:08:55 +0000 (03:08 +0000)]
bnx2x: fix select_queue when FCoE is disabled
Signed-off-by: Dmitry Kravkov <dmitry@broadcom.com> Signed-off-by: Eilon Greenstein <eilong@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
(cherry picked from commit cdb9d6ae8d128cc01d7a0649201102cd7de356e0)
Signed-off-by: Dmitry Kravkov <dmitry@broadcom.com> Signed-off-by: Eilon Greenstein <eilong@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
(cherry picked from commit f99779035730f8c3be8aeadd386cc9e77e086446)
Change XMAC loopback type from CORE LOCAL to LINE LOCAL for the BCM578xx due to intermittent problem with the loopback with this configuration.
Signed-off-by: Yaniv Rosner <yanivr@broadcom.com> Signed-off-by: Eilon Greenstein <eilong@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
(cherry picked from commit 4d7e25d6cc4312b1f949123fea7026fd56441513)
Vladislav Zolotarov [Tue, 9 Aug 2011 03:08:09 +0000 (03:08 +0000)]
bnx2x: init FCOE FP only once
Signed-off-by: Dmitry Kravkov <dmitry@broadcom.com> Signed-off-by: Eilon Greenstein <eilong@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
(cherry picked from commit 45d3539a2309858906abd8261c26f8ba8e50405a)
Remove remote fault detection as a tactic retreat due to link issues involved with it.
Once issue is resolved, this feature will be restored again.
Signed-off-by: Yaniv Rosner <yanivr@broadcom.com> Signed-off-by: Eilon Greenstein <eilong@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
(cherry picked from commit 0582242049c67d59c3a95cd1cba8995fa955c858)
Michael Chan [Mon, 20 Feb 2012 09:59:10 +0000 (09:59 +0000)]
cnic: update for FW 7.2.xx
(cherry picked from commit e65de0716f4fb72b09fb37f603b71f4eabe18322) Signed-off-by: Michael Chan <mchan@broadcom.com> Signed-off-by: Dmitry Kravkov <dmitry@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Joe Jin <joe.jin@oracle.com>
Commit 621b4d6 updated the bnx2x driver to a new FW version, but lacked
a commit to a header file with changes to the firmware's interface.
The missing interface change causes iscsi and fcoe to misbehave with the
updated firmware.
(cherry picked from commit f2ed5ee1b050495be49e5a0d5df152663558ef08) Signed-off-by: Yuval Mintz <yuvalmin@broadcom.com> Signed-off-by: Dmitry Kravkov <dmitry@broadcom.com> Signed-off-by: Eilon Greenstein <eilong@broadcom.com> CC: Michael Chan <mchan@broadcom.com> CC: Bhanu Prakash Gollapudi <bprakash@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Joe Jin <joe.jin@oracle.com>
Dmitry Kravkov [Mon, 26 Mar 2012 21:08:55 +0000 (21:08 +0000)]
bnx2x: fix vector traveling while looking for an empty entry
Fixes the bug that may prevent from mac to be configured,
while there is an empty slot for it.
Reported-by: Maciej Żenczykowski <zenczykowski@gmail.com>
(cherry picked from commit c54e9bd38a06babf94fd45e5f1df9a1109e12818) Signed-off-by: Dmitry Kravkov <dmitry@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Joe Jin <joe.jin@oracle.com>
Yuval Mintz [Mon, 12 Mar 2012 08:53:11 +0000 (08:53 +0000)]
bnx2x: mark functions as loaded on shared memory
This change enables the FW to make more accurate decisions regarding the
active functions.
(cherry picked from commit 9ce392d4fa43c3abb9ab02a29e225084ed84451e) Signed-off-by: Yuval Mintz <yuvalmin@broadcom.com> Signed-off-by: Eilon Greenstein <eilong@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Joe Jin <joe.jin@oracle.com>
Michal Schmidt [Thu, 15 Mar 2012 14:08:29 +0000 (14:08 +0000)]
bnx2x: fix memory leak in bnx2x_init_firmware()
When cycling the interface down and up, bnx2x_init_firmware() knows that
the firmware is already loaded, but nevertheless it allocates certain
arrays anew (init_data, init_ops, init_ops_offsets, iro_arr). The old
arrays are leaked.
Fix the leaks by returning early if the firmware was already loaded.
Because if the firmware is loaded, so are the arrays.
(cherry picked from commit c0ea452e422a1fc78ec8c639df64012d0b8dbb4a) Signed-off-by: Michal Schmidt <mschmidt@redhat.com> Acked-by: Eilon Greenstein <eilong@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Joe Jin <joe.jin@oracle.com>
Michal Schmidt [Thu, 15 Mar 2012 14:08:28 +0000 (14:08 +0000)]
bnx2x: fix a crash on corrupt firmware file
If the requested firmware is deemed corrupt and then released, reset the
pointer to NULL in order to avoid double-freeing it in
bnx2x_release_firmware() or dereferencing it in bnx2x_init_firmware().
(cherry picked from commit 127d0a198a310970b31866af8bbb6d4b1068e546) Signed-off-by: Michal Schmidt <mschmidt@redhat.com> Acked-by: Eilon Greenstein <eilong@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Joe Jin <joe.jin@oracle.com>
Michal Schmidt [Thu, 16 Feb 2012 02:38:48 +0000 (02:38 +0000)]
bnx2x: make bnx2x_close() static again
Commit 8304859a "bnx2x: add fan failure event handling" made the function
bnx2x_close() non-static unnecessarily. The function is not called from
other sources. Make it static again.
(cherry picked from commit 56ad315250bec1cd239c18712e072d1ff1f4aaf3) Signed-off-by: Michal Schmidt <mschmidt@redhat.com> Acked-by: Yuval Mintz <yuvalmin@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Joe Jin <joe.jin@oracle.com>
Mintz Yuval [Wed, 15 Feb 2012 02:10:31 +0000 (02:10 +0000)]
bnx2x: removed code re-checking memory base after device open
(cherry picked from commit 04f2d51313938d004983652ea6162d4940d83a04) Signed-off-by: Yuval Mintz <yuvalmin@broadcom.com> Signed-off-by: Eilon Greenstein <eilong@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Joe Jin <joe.jin@oracle.com>
Mintz Yuval [Wed, 15 Feb 2012 02:10:30 +0000 (02:10 +0000)]
bnx2x: allow BCM84833 phy to advertise 100Base-T speeds
(cherry picked from commit 10bd1f243c3a5ea1aebae88d21932e5614afa9cc) Signed-off-by: Yuval Mintz <yuvalmin@broadcom.com> Signed-off-by: Eilon Greenstein <eilong@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Joe Jin <joe.jin@oracle.com>
Mintz Yuval [Wed, 15 Feb 2012 02:10:29 +0000 (02:10 +0000)]
bnx2x: notify cnic of address of info-to-the-mcp
(cherry picked from commit 82fa848c117e57f0b3fd848f2aaf3f6859f42201) Signed-off-by: Yuval Mintz <yuvalmin@broadcom.com> Signed-off-by: Eilon Greenstein <eilong@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Joe Jin <joe.jin@oracle.com>
Mintz Yuval [Wed, 15 Feb 2012 02:10:28 +0000 (02:10 +0000)]
bnx2x: allocate smaller Rx rings for 1G functions
(cherry picked from commit d760fc37b0f74502b3f748951f22c6683b079a8e) Signed-off-by: Yuval Mintz <yuvalmin@broadcom.com> Signed-off-by: Eilon Greenstein <eilong@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Joe Jin <joe.jin@oracle.com>
Merav Sicron [Wed, 15 Feb 2012 02:10:27 +0000 (02:10 +0000)]
bnx2x: allocate memory dynamically in ethtool self-test.
Current ethtool self tests usesa large buffer on stack. This patch replaces
that array by dynamically allocated memory
(cherry picked from commit afa13b4b94bc4b3247fa46dd8698c8dbfe1a615c) Signed-off-by: Merav Sicron <meravs@broadcom.com> Signed-off-by: Eilon Greenstein <eilong@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Joe Jin <joe.jin@oracle.com>
Yaniv Rosner [Wed, 15 Feb 2012 02:10:26 +0000 (02:10 +0000)]
bnx2x: force 10G on 84833 phy should be autoneg with only 10G advertised
(cherry picked from commit 817a8aa8cfa1ebfb7bf27afb53ebb1ec3d49b564) Signed-off-by: Yaniv Rosner <yaniv.rosner@broadcom.com> Signed-off-by: Eilon Greenstein <eilong@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Joe Jin <joe.jin@oracle.com>
Yaniv Rosner [Wed, 15 Feb 2012 02:10:25 +0000 (02:10 +0000)]
bnx2x: added autoneg-restart after link advertisement changes
(cherry picked from commit 1b85ae52733b859e280035330e016731d4a62307) Signed-off-by: Yaniv Rosner <yaniv.rosner@broadcom.com> Signed-off-by: Eilon Greenstein <eilong@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Joe Jin <joe.jin@oracle.com>
Mintz Yuval [Wed, 15 Feb 2012 02:10:24 +0000 (02:10 +0000)]
bnx2x: ethtool publishes link partners speed and FC
Following the changes in the ethtool source code, this patch enables
the bnx2x driver to publish the Link partner's capabilities s, when ethtool
is used on an interface which completed autoneg.
(cherry picked from commit 9e7e8399c5d3c4dfaf84324a4a6b07a701d3e482) Signed-off-by: Yuval Mintz <yuvalmin@broadcom.com> Signed-off-by: Yaniv Rosner <yaniv.rosner@broadcom.com> Signed-off-by: Eilon Greenstein <eilong@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Joe Jin <joe.jin@oracle.com>
Yaniv Rosner [Wed, 15 Feb 2012 02:10:23 +0000 (02:10 +0000)]
bnx2x: half duplex support added for several boards
Several boards require an additional HW bit written in-order to enable
half duplex.
(cherry picked from commit e18c56b2e94080982d4542987a6fcf80b12d9414) Signed-off-by: Yaniv Rosner <yaniv.rosner@broadcom.com> Signed-off-by: Eilon Greenstein <eilong@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Joe Jin <joe.jin@oracle.com>
Michal Schmidt [Tue, 14 Feb 2012 09:05:46 +0000 (09:05 +0000)]
bnx2x: remove the 'poll' module option
'poll' was a debugging option, but turning it on these days leads to
kernel panic. Remove it.
(cherry picked from commit fc543637525b59af38af2ce09a4dbdd7d5eb27bf) Signed-off-by: Michal Schmidt <mschmidt@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Joe Jin <joe.jin@oracle.com>
Eric Dumazet [Mon, 13 Feb 2012 06:23:12 +0000 (06:23 +0000)]
bnx2x: fix bnx2x_storm_stats_update() on big endian
commit 619c5cb6885 (New 7.0 FW: bnx2x, cnic, bnx2i, bnx2fc) added new
sparse warnings.
(cherry picked from commit 66d885cba670059396b2f9ed9d5f4cbead0baee0) Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Cc: Eilon Greenstein <eilong@broadcom.com> Cc: Vladislav Zolotarov <vladz@broadcom.com> Cc: Michael Chan <mchan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Joe Jin <joe.jin@oracle.com>
Jesper Juhl [Mon, 6 Feb 2012 11:28:21 +0000 (11:28 +0000)]
bnx2x: Fix mem leak in bnx2x_tpa_stop() if build_skb() fails.
We allocate memory for 'new_data' with kmalloc(). If we get the memory
we then try to build_skb() and if that should fail (which it can) we
do not enter 'if (likely(skb)) {' and actually use 'new_data' but
instead fall through to the 'drop:' label and end up returning from
the function without ever assigning 'new'data' to anything or freeing
it. That leaks the memory allocated to 'new_data'.
This patch fixes the memory leak by doing a kfree(new_data) in the
case where build_skb() fails (or where allocation of 'new_data' itself
fails, but in taht case it's just a harmless kfree(NULL)).
(cherry picked from commit 3f61cd879c2f112c468e8849949b6fc88c739679) Signed-off-by: Jesper Juhl <jj@chaosbits.net> Acked-by: Eric Dumazet <eric.dumazet@gmail.com> Acked-by: Eilon Greenstein <eilong@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Joe Jin <joe.jin@oracle.com>
projects / users / jedix / linux-maple.git / log