virtio_net uses a timer to free old transmitted packets, rather than
leaving callbacks enabled all the time. If the host promises to
always notify us when the transmit ring is empty, we can free packets
at that point and avoid the timer.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
virtio_net currently only frees old transmit skbs just
before queueing new ones. If the queue is full, it then
enables interrupts and waits for notification that more
work has been performed.
However, a side-effect of this scheme is that there are
always xmit skbs left dangling when no new packets are
sent, against the Documentation/networking/driver.txt
guideline:
"... it is not allowed for your TX mitigation scheme
to let TX packets "hang out" in the TX ring unreclaimed
forever if no new TX packets are sent."
Add a timer to ensure that any time we queue new TX
skbs, we will shortly free them again.
This fixes an easily reproduced hang at shutdown where
iptables attempts to unload nf_conntrack and nf_conntrack
waits for an skb it is tracking to be freed, but virtio_net
never frees it.
Signed-off-by: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
Signed-off-by: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
hdr->csum_start is the offset from the start of the ethernet
header to the transport layer checksum field. skb->csum_start
is the offset from skb->head.
skb_partial_csum_set() assumes that skb->data points to the
ethernet header - i.e. it computes skb->csum_start by adding
the headroom to hdr->csum_start.
Since eth_type_trans() skb_pull()s the ethernet header,
skb_partial_csum_set() should be called before
eth_type_trans().
(Without this patch, GSO packets from a guest to the world outside the
host are corrupted).
Signed-off-by: Mark McLoughlin <markmc@redhat.com>
Acked-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
eHEA has to call firmware functions in order to change the mac address
of a logical port. This patch checks if the logical port is up
when calling the register / deregister mac address calls. If the port
is down these firmware calls would fail and are therefore not executed.
Signed-off-by: Jan-Bernd Themann <themann@de.ibm.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
RX queue flush can fail if traffic continues to arrive. Recover by
performing an invisible reset.
Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
Add a workaround for lost MSI interrupts. There is a race condition in
the HW in which future interrupts could be missed. The workaround is to
toggle the MSI irq mask.
Added cleanup based on comments from Andrew Morton.
Signed-off-by: Ayaz Abdulla <aabdulla@nvidia.com>
Cc: Manfred Spraul <manfred@colorfullife.com>
Cc: Jeff Garzik <jeff@garzik.org>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
When pfkey has no km listeners, it still does a lot of work
before finding out there aint nobody out there.
If a tree falls in a forest and no one is around to hear it, does it make
a sound? In this case it makes a lot of noise:
With this short-circuit adding 10s of thousands of SAs using
netlink improves performance by ~10%.
Signed-off-by: Jamal Hadi Salim <hadi@cyberus.ca>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch fixes several issues:
Use the right openprom device name so the driver is actually loaded.
Fix a crash due to unitialized info->pseudo_palette.
Put the framebuffer in the proper mode for software rendering.
checkpatch cleanups.
Hardware acceleration was removed when the driver was rewritten
for the new framebuffer API in 2003. Software rendering requires
a different framebuffer access mode but that wasn't changed. The
driver now works again but is slow. The proper fix is to reintroduce
hardware acceleration.
Signed-off-by: Robert Reif <reif@earthlink.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
We need to unshare the skb first as otherwise pskb_may_pull may
write to a shared skb which could be bad.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
The length field in the PPPOE header wasn't checked completely.
This patch causes all packets shorter than the declared length
to be dropped.
It also changes the memcpy_toiovec call to skb_copy_datagram_iovec
so that paged packets (rare for PPPOE) are handled properly.
Thanks to Ilja of the Netric Security Team for discovering and
reporting this bug, and Chris Wright for the total_len check.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Acked-by: Karsten Keil <kkeil@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
1. creating proc entry and not saving pointer to PDE and checking it
is not going to work.
2. if proc entry wasn't created, no reason to remove it on error path.
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
KERNEL_OFFSET macro in eni.h is not required as it is not used anywhere.
Remove the unused macro from eni.h header file.
Signed-off-by: Pradeep Singh <rautelap@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Wei Yongjun noticed that we may call reqsk_free on request sock objects where
the opt fields may not be initialized, fix it by introducing inet_reqsk_alloc
where we initialize ->opt to NULL and set ->pktopts to NULL in
inet6_reqsk_alloc.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
- The tcp_unhash() method in /include/net/tcp.h is no more needed, as the
unhash method in tcp_prot structure is now inet_unhash (instead of
tcp_unhash in the
past); see tcp_prot structure in net/ipv4/tcp_ipv4.c.
- So, this patch removes tcp_unhash() declaration from include/net/tcp.h
Signed-off-by: Rami Rosen <ramirose@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch fixes a potential memory corruption in
pppol2tp_recvmsg(). If skb->len is bigger than the caller's buffer
length, memcpy_toiovec() will go into unintialized data on the kernel
heap, interpret it as an iovec and start modifying memory.
The fix is to change the memcpy_toiovec() call to
skb_copy_datagram_iovec() so that paged packets (rare for PPPOL2TP)
are handled properly. Also check that the caller's buffer is big
enough for the data and set the MSG_TRUNC flag if it is not so.
Reported-by: Ilja <ilja@netric.org>
Signed-off-by: James Chapman <jchapman@katalix.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch limits BLK_DEV_HD_ONLY to the ARM platforms offering
IRQ_HARDDISK, fixing the following compile error on others:
<-- snip -->
...
CC drivers/ide/legacy/hd.o
...
/home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/ide/legacy/hd.c: In function 'hd_times_out':
/home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/ide/legacy/hd.c:542: error: 'IRQ_HARDDISK' undeclared (first use in this function)
/home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/ide/legacy/hd.c:542: error: (Each undeclared identifier is reported only once
/home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/ide/legacy/hd.c:542: error: for each function it appears in.)
/home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/ide/legacy/hd.c: In function 'do_hd_request':
/home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/ide/legacy/hd.c:661: error: 'IRQ_HARDDISK' undeclared (first use in this function)
/home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/ide/legacy/hd.c: In function 'hd_init':
/home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/ide/legacy/hd.c:765: error: 'IRQ_HARDDISK' undeclared (first use in this function)
make[3]: *** [drivers/ide/legacy/hd.o] Error 1
<-- snip -->
Reported-by: Adrian Bunk <bunk@kernel.org>
Cc: Russell King <rmk@arm.linux.org.uk>
Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Don't fail the probe if there are no devices attached to the controller.
Cc: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Convert the driver to use struct ide_port_info - as a nice side-effect
this fixes racy setup of ->io_32bit/unmask settings (after ide_device_add()
call device can be already in use).
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
hwif->dev was set too late (after ide_device_add() call)
so hwif->gendev.parent was not initialized properly.
Fix it by setting hw.dev and letting ide_init_port_hw()
do the rest.
Acked-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
ide_find_port() now depends on ->chipset being set for occupied ide_hwifs[]
slots so all host drivers have to initialize hwif->chipset properly.
This patch fixes a regression on hosts with > 1 port or with a single port
but no devices attached to it for an affected host drivers.
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
hwif->chipset need to be set properly or ide-generic driver will break once
we make a final step in fixing host drivers' dependence on ide_hwifs[].
Problem was catched early thanks to IDE tree exposure in -mm / -next trees
and reported by people listed people (thank you guys!).
Reported-by: "John Keller" <jpk@sgi.com>
Reported-by: Dmitri Vorobiev <dmitri.vorobiev@gmail.com>
Reported-by: Mel Gorman <mel@csn.ul.ie>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
- maintainer has not been active for years
- URLs no longer exist
- covered by the IDE SUBSYSTEM entry
- maintainer email bounces
Signed-off-by: Adrian Bunk <bunk@kernel.org>
Cc: Lionel.Bouton@inet6.fr
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
I forgot to remove the ide_etrax100 chipset type when removing the
ETRAX_IDE driver.
Reported-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Found a silly double assignment of err is do_shmat. Silly, but good to
clean up the useless code.
Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
schedule() has the special "TASK_INTERRUPTIBLE && signal_pending()" case,
this allows us to do
current->state = TASK_INTERRUPTIBLE;
schedule();
without fear to sleep with pending signal.
However, the code like
current->state = TASK_KILLABLE;
schedule();
is not right, schedule() doesn't take TASK_WAKEKILL into account. This means
that mutex_lock_killable(), wait_for_completion_killable(), down_killable(),
schedule_timeout_killable() can miss SIGKILL (and btw the second SIGKILL has
no effect).
Introduce the new helper, signal_pending_state(), and change schedule() to
use it. Hopefully it will have more users, that is why the task's state is
passed separately.
Note this "__TASK_STOPPED | __TASK_TRACED" check in signal_pending_state().
This is needed to preserve the current behaviour (ptrace_notify). I hope
this check will be removed soon, but this (afaics good) change needs the
separate discussion.
The fast path is "(state & (INTERRUPTIBLE | WAKEKILL)) + signal_pending(p)",
basically the same that schedule() does now. However, this patch of course
bloats schedule().
Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
The semantic match that finds the problem is as follows:
(http://www.emn.fr/x-info/coccinelle/)
@r exists@
expression E,E1;
statement S;
position p1,p2,p3;
@@
E =@p1 \(kmalloc\|kcalloc\|kzalloc\)(...)
... when != E = E1
if (E == NULL || ...) S
... when != E = E1
if@p2 (...) {
... when != kfree(E)
}
... when != E = E1
kfree@p3(E);
@forall@
position r.p2;
expression r.E;
int E1 != 0;
@@
* if@p2 (...) {
... when != kfree(E)
when strict
return E1; }
Signed-off-by: Julia Lawall <julia@diku.dk>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
__ctl_load/__ctl_store are called with either an array of unsigned long or
a single unsigned long value. Add an address operator to the "m"/"=m"
contraints to make them work for unsigned long arguments as well.
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
The first argument to __ctl_store() should be the array to store
stuff in, not just the first element of that array. With the
current code in __cpu_up(), mainline GCC dies with an internal
compiler error. I didn't diagnose that further, but just fixed
the kernel bug.
Signed-off-by: Segher Boessenkool <segher@kernel.crashing.org>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Priorities for I/O interruption subclasses range from 0 (highest)
to 7 (lowest). Unfortunately, the console has been using isc 7
instead of an isc with a higher priority than regular I/O
subchannels (which use 3). Fix this by making the console use
isc 1.
Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
If a memory range is supposed to be added to the 1:1 mapping and it
ends just below the maximum supported physical address it won't
succeed. This is because a test doesn't consider that the end address
is 1 smaller than start + size.
Fix the comparison.
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
sparse complains about signedness:
drivers/s390/cio/blacklist.c:132:28: warning: incorrect type in
argument 2 (different signedness)
drivers/s390/cio/blacklist.c:132:28: expected unsigned int *val
drivers/s390/cio/blacklist.c:132:28: got int *cssid
drivers/s390/cio/blacklist.c:136:28: warning: incorrect type in
argument 2 (different signedness)
drivers/s390/cio/blacklist.c:136:28: expected unsigned int *val
drivers/s390/cio/blacklist.c:136:28: got int *ssid
drivers/s390/cio/blacklist.c:140:28: warning: incorrect type in
argument 2 (different signedness)
drivers/s390/cio/blacklist.c:140:28: expected unsigned int *val
drivers/s390/cio/blacklist.c:140:28: got int *devno
cssid, ssid and devno are of course unsigned, so let's make the
variables unsigned as well.
Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
In case of !64BIT kernel we end up with a zero sized mem_section array.
This happens because NR_MEM_SECTIONS is smaller than SECTIONS_PER_ROOT
but we have:
#define NR_SECTION_ROOTS (NR_MEM_SECTIONS / SECTIONS_PER_ROOT)
and
struct mem_section *mem_section[NR_SECTION_ROOTS];
So fix this by selecting SPARSEMEM_STATIC which makes sure
that SECTIONS_PER_ROOT is 1.
Cc: Gerald Schaefer <gerald.schaefer@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband:
IB/core: Remove IB_DEVICE_SEND_W_INV capability flag
IB/umem: Avoid sign problems when demoting npages to integer
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/cooloney/blackfin-2.6:
Blackfin serial driver: fix up tty core set_ldisc API change breakage bug
Blackfin arch: protect only the SPI bus controller with CONFIG_SPI_BFIN
Blackfin arch: fixup warnings with the new cplb saved values
Blackfin Serial Driver: Clean up BF54x macro in blackfin UART driver.
The bindv6only is tuned via sysctl. It is already on a struct net
and per-net sysctls allow for its modification (ipv6_sysctl_net_init).
Despite this the value configured in the init net is used for the
rest of them.
Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Rusty Russell