There are some fields in "edata" which have not been cleared. One
example is edata.cmd. It leaks uninitialized stack information to the
user.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Eliminate some warnings by casting to unsigned long before
casting a dma_addr_t value to a pointer.
btw:
Does slicoss always work on x86-32?
Is a pshmem guaranteed to be accessible
by a 32 bit address?
Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Instead of performing the hash calculation for the mac address by ourself, we
can simply reuse ether_crc and shift only the result according to our
needs.
The code was tested against the previous implementation by verifying both
implementations against each other in userspace for 16200000000 different
mac addresses, changing the vendor bits of the mac address first.
Signed-off-by: Peter Huewe <peterhuewe@gmx.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
smatch complains about a possible buffer overflow
slicoss.c:3651 slic_card_locate() error: buffer overflow
'physcard->adapter' 4 <= 4
If the for loop is not exited prematurely i++ is executed after the last
iteration and thus i can be 4, which is out of bounds for
physcard->adapter.
-> Add check for this condition and simplify the if statement by
inverting the condition.
Signed-off-by: Peter Huewe <peterhuewe@gmx.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
gcc complains about an undefined operation:
slicoss.c:1417:19: warning: operation on 'rspq->pageindex' may be
undefined [-Wsequence-point]
The intended operation was (probably) to retrieve the pageindex + 1 and let
it wrap around if it reaches the num_pages.
Signed-off-by: Peter Huewe <peterhuewe@gmx.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
skbtype is assigned once to NORMAL_ETHFRAME and then checked if it is
NORMAL_ETHFRAME -> remove the checks.
This also gets rid of the (false positive) smatch warning:
slicoss.c:2829 slic_xmit_start() error: potential NULL dereference
'hcmd'.
Signed-off-by: Peter Huewe <peterhuewe@gmx.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Smatch complains that the variable adapter is dereferenced before it is
checked:
slicoss.c:906 slic_timer_load_check() warn: variable dereferenced before
check 'adapter' (see line 904)
-> move the assignment after the check.
Signed-off-by: Peter Huewe <peterhuewe@gmx.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This fixes all instances of
"Please, no spaces at start of a new line"
"Please, no spaces before tabs"
Please note that I probably got the warning names wrong,
but they should be close enough for usage here :)
Additional post-commit note: There is one comment on line 230ish in slic.h
that appears to have lost it's formatting. It was fine when I was working
in Geany, but it caught my eye in the below diff. Sorry if it actually happened!
Signed-off-by: Cruz Julian Bishop <cruzjbishop@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The errors fixed are all instances of
"ERROR: do not use assignment in if condition"
Signed-off-by: Cruz Julian Bishop <cruzjbishop@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
I'm assuming that "CRC shit reg" is referencing the CRC register/registry.
Please ignore this patch if "shit" is actually in context
Signed-off-by: Cruz Julian Bishop <cruzjbishop@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
CONFIG_HOTPLUG is going away as an option so __devexit is no
longer needed.
Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
CONFIG_HOTPLUG is going away as an option so __devinit is no longer
needed.
Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
CONFIG_HOTPLUG is going away as an option so __devexit_p is no longer
needed.
Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
slic_card_locate does a kzalloc of physcard, and if it fails,
in my previous patch i returned -ENOMEM, but left the driver leak
the memory if card_hostid == SLIC_HOSTID_DEFAULT,
fix this memory leak if the above condition is true
Signed-off-by: Devendra Naga <devendra.aaru@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
this function memset's the cmdqmem, instead do it in slic_cmdq_init.
Signed-off-by: Devendra Naga <devendra.aaru@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
this default statement does nothing in the irqhandler, so remove it
Signed-off-by: Devendra Naga <devendra.aaru@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
this takes up the error path cleanup,
fixes a crash too due to null deref
Signed-off-by: Devendra Naga <devendra.aaru@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
remove ASSERT and remove all its callers altogether in the code
Signed-off-by: Devendra Naga <devendra.aaru@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
we are dereferencing the pshmem , and the pci_alloc_consistent can fail
returning null,
do a memcpy if we have a valid pshmem
Signed-off-by: Devendra Naga <devendra.aaru@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
this function return void, means return at the end of this function
is not needed
Signed-off-by: Devendra Naga <devendra.aaru@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
the locked variable is used for checking whether the function
acquired lock, then unlock. actually with out this we
can achieve the same lock and unlock senario, remove the locked
variable and also cleanup the code around.
Signed-off-by: Devendra Naga <devendra.aaru@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
we are returning -ENOENT when there is no firmware file for a matching
device id. then we start calling request_firmware,
after this we do checks on the firmware length of corresponding
device id, since the default case is handled in the begining itself
there is no need of a default case at the firmware length checks
Signed-off-by: Devendra Naga <devendra.aaru@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
we request_firmware in slic_card_download_gbrcv and we return out
with out calling release_firmware, where we compare against a
firmware lengths of certain device ids.
Signed-off-by: Devendra Naga <devendra.aaru@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
when the device id doesn't match in slic_card_download_gbrcv the filename is "",
i.e an empty name, and we try calling request_firmware with that name,
actually we can just fail out at default case before even calling request_firmware
Signed-off-by: Devendra Naga <devendra.aaru@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
at probe we enabled the device, and we should disable it at remove.
Signed-off-by: Devendra Naga <devendra.aaru@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
if no case matches we are simply asserting and doing break.
and i think we may need to return that -ENODEV , no device is
present, rather assert'ing.
Signed-off-by: Devendra Naga <devendra.aaru@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
As the private pointer is valid at the remove of driver, and remove wont' be called if probe fails, so no point for checking of ASSERT
Signed-off-by: Devendra Naga <devendra.aaru@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The below patch fixes a typo found while reading through staging/slicoss
Signed-off-by: Justin P. Mattock <justinmattock@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Mark Einon points out that the Kconfig option for NETDEV_1000 no longer
exists, and the merge of the staging drivers should have removed that
for the et131x driver.
And while checking for it, I noticed that slicoss had the same stale
dependency. Remove that one too.
Reported-by: Mark Einon <mark.einon@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Freeing netdev without free_netdev() leads to net, tx leaks.
I might lead to dereferencing freed pointer.
The semantic match that finds this problem is as follows:
(http://coccinelle.lip6.fr/)
@@
struct net_device* dev;
@@
-kfree(dev)
+free_netdev(dev)
Signed-off-by: Vasiliy Kulikov <segooon@gmail.com>
Acked-by: Denis Kirjanov <dkirjanov@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This patch makes error handling more readable due to 'goto err' pattern.
Signed-off-by: Kulikov Vasiliy <segooon@gmail.com>
Cc: Denis Kirjanov <kirjanov@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This patch makes error handling more readable due to 'goto err' pattern.
Signed-off-by: Kulikov Vasiliy <segooon@gmail.com>
Cc: Denis Kirjanov <kirjanov@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Converts the list and the core manipulating with it to be the same as uc_list.
+uses two functions for adding/removing mc address (normal and "global"
variant) instead of a function parameter.
+removes dev_mcast.c completely.
+exposes netdev_hw_addr_list_* macros along with __hw_addr_* functions for
manipulation with lists on a sandbox (used in bonding and 80211 drivers)
Signed-off-by: Jiri Pirko <jpirko@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
There were a number of patches that went into Linus's
tree already that conflicted with other changes in the
staging branch. This merge resolves those merge conflicts.
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
String constants that are continued on subsequent lines with \
are not good.
Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Use pci_set_consistent_dma_mask() in the case of 64-bit consistent allocations.
Signed-off-by: Denis Kirjanov <kirjanov@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Use PCI_DEVICE_TABLE: defines array as const and puts
it into the __devinitconst section.
Signed-off-by: Denis Kirjanov <kirjanov@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Dan Carpenter