| Age | Commit message (Collapse) | Author |
|---|
|
rebased
Rebased on btrfs-next and retested.
Inform should_defrag_range if BTRFS_DEFRAG_RANGE_COMPRESS is set. If so, skip
checks for adjacent extents and extent size when deciding whether to defrag,
as these can prevent an uncompressed and unfragmented file from being
compressed as requested.
Signed-off-by: Andrew Mahone <andrew.mahone@gmail.com>
|
|
"root->fs_info" and "fs_info" are the same, but "fs_info" is prefered
because it is shorter and that's what is used in the rest of the
function.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
|
|
Before the update_time inode operation was indroduced, it was
not possible to prevent updates of atime on RO subvolumes. VFS
was only able to check for RO on the mount, but did not know
anything about btrfs subvolumes.
btrfs_update_time does now check if the root is RO and skip
updating of times.
Signed-off-by: Alexander Block <ablock84@googlemail.com>
|
|
Btrfs allows to turn on compression on a mounted and used filesystem
by issuing mount -o remount,compress=lzo.
This patch allows to turn compression off again
while the filesystem is mounted. As suggested by David Sterba
if the compress-force option was set, it is implicitly cleared
if compression is turned off.
Tested-by: David Sterba <dsterba@suse.cz>
Signed-off-by: Arnd Hannemann <arnd@arndnet.de>
|
|
We do all of our inode updating when we change it, and now that we do
->update_time we don't need ->dirty_inode for atime updates anymore, so just
remove it. Thanks,
Signed-off-by: Josef Bacik <josef@redhat.com>
|
|
The btrfs locks were unconditionally calling wake_up as the
locks were released. This lead to extra thrashing on the waitqueue,
especially for locks that were dominated by readers.
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
|
|
Waiting on spindles improves performance, but ssds want all the
IO as quickly as we can push it down.
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs
Pull the big VFS changes from Al Viro:
"This one is *big* and changes quite a few things around VFS. What's in there:
- the first of two really major architecture changes - death to open
intents.
The former is finally there; it was very long in making, but with
Miklos getting through really hard and messy final push in
fs/namei.c, we finally have it. Unlike his variant, this one
doesn't introduce struct opendata; what we have instead is
->atomic_open() taking preallocated struct file * and passing
everything via its fields.
Instead of returning struct file *, it returns -E... on error, 0
on success and 1 in "deal with it yourself" case (e.g. symlink
found on server, etc.).
See comments before fs/namei.c:atomic_open(). That made a lot of
goodies finally possible and quite a few are in that pile:
->lookup(), ->d_revalidate() and ->create() do not get struct
nameidata * anymore; ->lookup() and ->d_revalidate() get lookup
flags instead, ->create() gets "do we want it exclusive" flag.
With the introduction of new helper (kern_path_locked()) we are rid
of all struct nameidata instances outside of fs/namei.c; it's still
visible in namei.h, but not for long. Come the next cycle,
declaration will move either to fs/internal.h or to fs/namei.c
itself. [me, miklos, hch]
- The second major change: behaviour of final fput(). Now we have
__fput() done without any locks held by caller *and* not from deep
in call stack.
That obviously lifts a lot of constraints on the locking in there.
Moreover, it's legal now to call fput() from atomic contexts (which
has immediately simplified life for aio.c). We also don't need
anti-recursion logics in __scm_destroy() anymore.
There is a price, though - the damn thing has become partially
asynchronous. For fput() from normal process we are guaranteed
that pending __fput() will be done before the caller returns to
userland, exits or gets stopped for ptrace.
For kernel threads and atomic contexts it's done via
schedule_work(), so theoretically we might need a way to make sure
it's finished; so far only one such place had been found, but there
might be more.
There's flush_delayed_fput() (do all pending __fput()) and there's
__fput_sync() (fput() analog doing __fput() immediately). I hope
we won't need them often; see warnings in fs/file_table.c for
details. [me, based on task_work series from Oleg merged last
cycle]
- sync series from Jan
- large part of "death to sync_supers()" work from Artem; the only
bits missing here are exofs and ext4 ones. As far as I understand,
those are going via the exofs and ext4 trees resp.; once they are
in, we can put ->write_super() to the rest, along with the thread
calling it.
- preparatory bits from unionmount series (from dhowells).
- assorted cleanups and fixes all over the place, as usual.
This is not the last pile for this cycle; there's at least jlayton's
ESTALE work and fsfreeze series (the latter - in dire need of fixes,
so I'm not sure it'll make the cut this cycle). I'll probably throw
symlink/hardlink restrictions stuff from Kees into the next pile, too.
Plus there's a lot of misc patches I hadn't thrown into that one -
it's large enough as it is..."
* 'for-linus-2' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs: (127 commits)
ext4: switch EXT4_IOC_RESIZE_FS to mnt_want_write_file()
btrfs: switch btrfs_ioctl_balance() to mnt_want_write_file()
switch dentry_open() to struct path, make it grab references itself
spufs: shift dget/mntget towards dentry_open()
zoran: don't bother with struct file * in zoran_map
ecryptfs: don't reinvent the wheels, please - use struct completion
don't expose I_NEW inodes via dentry->d_inode
tidy up namei.c a bit
unobfuscate follow_up() a bit
ext3: pass custom EOF to generic_file_llseek_size()
ext4: use core vfs llseek code for dir seeks
vfs: allow custom EOF in generic_file_llseek code
vfs: Avoid unnecessary WB_SYNC_NONE writeback during sys_sync and reorder sync passes
vfs: Remove unnecessary flushing of block devices
vfs: Make sys_sync writeout also block device inodes
vfs: Create function for iterating over block devices
vfs: Reorder operations during sys_sync
quota: Move quota syncing to ->sync_fs method
quota: Split dquot_quota_sync() to writeback and cache flushing part
vfs: Move noop_backing_dev_info check from sync into writeback
...
|
|
Signed-off-by: Wanpeng Li <liwp.linux@gmail.com>
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
|
|
Signed-off-by: Sasha Levin <levinsasha928@gmail.com>
[v1: Rebased with tracing removed]
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
|
|
This was formerly used to store the tclk value. This is now discovered
using the clk API, rather than pass it as platform data.
Signed-off-by: Michael Walle <michael@walle.cc>
Acked-by: Jason Cooper <jason@lakedaemon.net>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Acked-by: Sebastian Hesselbarth <sebastian.hesselbarth@googlemail.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
|
|
We are currently using an inferior or equal operator for comparing
the transfer frequency with the clock frequency table. Because of
this, we always end up selecting 20Mhz as a frequency, due to the
inequality transfer hz <= 20 Mhz being always true. Fix this by
reversing the inequality, which is how the comparison should be done.
Signed-off-by: Florian Fainelli <florian@openwrt.org>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound into for-next
ASoC: Additional updates for 3.6
A few more fixes for 3.6, some of which are relatively important -
they've all been in -next for at least some time.
- DAPM fixes for the recent locking changes.
- Fix for _PRE and _POST widgets (which have been broken for a few
releases now).
- A couple of minor driver updates.
|
|
Convert orion_wdt driver to use watchdog framework API.
Signed-off-by: Axel Lin <axel.lin@gmail.com>
Tested-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
|
|
module_platform_driver() replaces module_init() and module_exit()
and makes the code simpler.
Signed-off-by: Sachin Kamat <sachin.kamat@linaro.org>
Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
|
|
Some motherboards like the Advantech ARK3400 documentation
use a non-inverted GPIO pin. We fix this by assuming that
the BIOS will set the Polarity bit for the GPIO correctly
at startup and we keep the Bit-setting intact when we start
and stop the watchdog.
Reported-by: Jean-François Deverge <jf.deverge@gmail.com>
Signed-off-by: Dave Mueller <d.mueller@elsoft.ch>
Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
|
|
intransition state.
OMAP watchdog driver is adapted to runtime PM like a general device
driver but it is not appropriate. It is causing couple of functional
issues.
1. On OMAP4 SYSCLK can't be gated, because of issue with WDTIMER2 module,
which constantly stays in "in transition" state. Value of register
CM_WKUP_WDTIMER2_CLKCTRL is always 0x00010000 in this case.
Issue occurs immediately after first idle, when hwmod framework tries
to disable WDTIMER2 functional clock - "wd_timer2_fck". After this
module falls to "in transition" state, and SYSCLK gating is blocked.
2. Due to runtime PM, watchdog timer may be completely disabled.
In current code base watchdog timer is not disabled only because of
issue 1. Otherwise state of WDTIMER2 module will be "Disabled", and there
will be no interrupts from omap_wdt. In other words watchdog will not
work at all.
Watchdong is a special IP and it should not be disabled otherwise
purpose of it itself is defeated. Watchdog functional clock should
never be disabled. This patch updates the runtime PM handling in
driver so that runtime PM is limited only during probe/shutdown
and suspend/resume.
The patch fixes issue 1 and 2
Signed-off-by: Lokesh Vutla <lokeshvutla@ti.com>
Acked-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
|
|
ie6xx_wdt_probe() calls ie6xx_wdt_debugfs_exit() as part of
it's error cleanup path, and ie6xx_wdt_debugfs_exit() is
currently annotated __devexit.
Signed-off-by: Gerard Snitselaar <dev@snitselaar.org>
Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
|
|
bcm63xx_wdt was used as a platform_driver but was not suffixed with
_driver, thus causing section mismatches, fix that.
Signed-off-by: Florian Fainelli <florian@openwrt.org>
Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
|
|
This patch converts the iTCO_wdt watchdog driver to use the
generic watchdog framework.
Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
|
|
This watchdog driver had ioctl defines introduced locally
for pre timeout handling, marked to be removed as soon as
a generic replacement would become available.
The latter has actually occurred in 2006, at e05b59fe.
Remove the local duplicates for pre timeout handling.
Signed-off-by: Oskar Schirmer <oskar@scara.com>
Acked-by: Corey Minyard <cminyard@mvista.com>
Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
|
|
Do not use clock-frequency property from parent node.
Use it from watchdog node.
Signed-off-by: Michal Simek <monstr@monstr.eu>
Acked-By: Alejandro Cabrera <acabrera@udio.cujae.edu.cu>
Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
|
|
Make sure we prepare/unprepare the COH901327 watchdog timer
as is required by the clk API especially if you use common
clock.
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Reviewed-by : Pankaj Jangra <jangra.pankaj9@gmail.com>
Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
|
|
The Jetway JNF99 motherboard features a F71869 SuperIO chip, but its
watchdog chipset ID appears to be 1007 (as opposed to 0814). Some testing
confirmed it behaves the exact same as 0814. So add this chipset ID to the
module's ID list so that the Fintek watchdog driver can correctly identify
and access it.
Signed-off-by: Justin Wheeler <jwheeler@datademons.com>
Acked-by: Giel van Schijndel <me@mortis.eu>
Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
|
|
Signed-off-by: Michael Walle <michael@walle.cc>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Acked-by: Jason Cooper <jason@lakedaemon.net>
Acked-by: Sebastian Hesselbarth <sebastian.hesselbarth@googlemail.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
|
|
'api/domain-attr' into next
Conflicts:
drivers/iommu/iommu.c
include/linux/iommu.h
|
|
Add Asus All-In-One PC keyboard model AK1D.
BugLink: https://bugs.launchpad.net/bugs/1027789
Signed-off-by: Cyrus Lien <cyrus.lien@canonical.com>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
|
|
|
|
Ever since the DAPM performance improvements we've been marking all widgets
as not dirty after each DAPM run. Since _PRE and _POST events aren't part
of the DAPM graph this has rendered them non-functional, they will never be
marked dirty again and thus will never be run again.
Fix this by skipping them when marking widgets as not dirty.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Acked-by: Liam Girdwood <lrg@ti.com>
Cc: stable@vger.kernel.org
|
|
Jay Fenlason (fenlason@redhat.com) found a bug,
that recvfrom() on an RDS socket can return the contents of random kernel
memory to userspace if it was called with a address length larger than
sizeof(struct sockaddr_in).
rds_recvmsg() also fails to set the addr_len paramater properly before
returning, but that's just a bug.
There are also a number of cases wher recvfrom() can return an entirely bogus
address. Anything in rds_recvmsg() that returns a non-negative value but does
not go through the "sin = (struct sockaddr_in *)msg->msg_name;" code path
at the end of the while(1) loop will return up to 128 bytes of kernel memory
to userspace.
And I write two test programs to reproduce this bug, you will see that in
rds_server, fromAddr will be overwritten and the following sock_fd will be
destroyed.
Yes, it is the programmer's fault to set msg_namelen incorrectly, but it is
better to make the kernel copy the real length of address to user space in
such case.
How to run the test programs ?
I test them on 32bit x86 system, 3.5.0-rc7.
1 compile
gcc -o rds_client rds_client.c
gcc -o rds_server rds_server.c
2 run ./rds_server on one console
3 run ./rds_client on another console
4 you will see something like:
server is waiting to receive data...
old socket fd=3
server received data from client:data from client
msg.msg_namelen=32
new socket fd=-1067277685
sendmsg()
: Bad file descriptor
/***************** rds_client.c ********************/
int main(void)
{
int sock_fd;
struct sockaddr_in serverAddr;
struct sockaddr_in toAddr;
char recvBuffer[128] = "data from client";
struct msghdr msg;
struct iovec iov;
sock_fd = socket(AF_RDS, SOCK_SEQPACKET, 0);
if (sock_fd < 0) {
perror("create socket error\n");
exit(1);
}
memset(&serverAddr, 0, sizeof(serverAddr));
serverAddr.sin_family = AF_INET;
serverAddr.sin_addr.s_addr = inet_addr("127.0.0.1");
serverAddr.sin_port = htons(4001);
if (bind(sock_fd, (struct sockaddr*)&serverAddr, sizeof(serverAddr)) < 0) {
perror("bind() error\n");
close(sock_fd);
exit(1);
}
memset(&toAddr, 0, sizeof(toAddr));
toAddr.sin_family = AF_INET;
toAddr.sin_addr.s_addr = inet_addr("127.0.0.1");
toAddr.sin_port = htons(4000);
msg.msg_name = &toAddr;
msg.msg_namelen = sizeof(toAddr);
msg.msg_iov = &iov;
msg.msg_iovlen = 1;
msg.msg_iov->iov_base = recvBuffer;
msg.msg_iov->iov_len = strlen(recvBuffer) + 1;
msg.msg_control = 0;
msg.msg_controllen = 0;
msg.msg_flags = 0;
if (sendmsg(sock_fd, &msg, 0) == -1) {
perror("sendto() error\n");
close(sock_fd);
exit(1);
}
printf("client send data:%s\n", recvBuffer);
memset(recvBuffer, '\0', 128);
msg.msg_name = &toAddr;
msg.msg_namelen = sizeof(toAddr);
msg.msg_iov = &iov;
msg.msg_iovlen = 1;
msg.msg_iov->iov_base = recvBuffer;
msg.msg_iov->iov_len = 128;
msg.msg_control = 0;
msg.msg_controllen = 0;
msg.msg_flags = 0;
if (recvmsg(sock_fd, &msg, 0) == -1) {
perror("recvmsg() error\n");
close(sock_fd);
exit(1);
}
printf("receive data from server:%s\n", recvBuffer);
close(sock_fd);
return 0;
}
/***************** rds_server.c ********************/
int main(void)
{
struct sockaddr_in fromAddr;
int sock_fd;
struct sockaddr_in serverAddr;
unsigned int addrLen;
char recvBuffer[128];
struct msghdr msg;
struct iovec iov;
sock_fd = socket(AF_RDS, SOCK_SEQPACKET, 0);
if(sock_fd < 0) {
perror("create socket error\n");
exit(0);
}
memset(&serverAddr, 0, sizeof(serverAddr));
serverAddr.sin_family = AF_INET;
serverAddr.sin_addr.s_addr = inet_addr("127.0.0.1");
serverAddr.sin_port = htons(4000);
if (bind(sock_fd, (struct sockaddr*)&serverAddr, sizeof(serverAddr)) < 0) {
perror("bind error\n");
close(sock_fd);
exit(1);
}
printf("server is waiting to receive data...\n");
msg.msg_name = &fromAddr;
/*
* I add 16 to sizeof(fromAddr), ie 32,
* and pay attention to the definition of fromAddr,
* recvmsg() will overwrite sock_fd,
* since kernel will copy 32 bytes to userspace.
*
* If you just use sizeof(fromAddr), it works fine.
* */
msg.msg_namelen = sizeof(fromAddr) + 16;
/* msg.msg_namelen = sizeof(fromAddr); */
msg.msg_iov = &iov;
msg.msg_iovlen = 1;
msg.msg_iov->iov_base = recvBuffer;
msg.msg_iov->iov_len = 128;
msg.msg_control = 0;
msg.msg_controllen = 0;
msg.msg_flags = 0;
while (1) {
printf("old socket fd=%d\n", sock_fd);
if (recvmsg(sock_fd, &msg, 0) == -1) {
perror("recvmsg() error\n");
close(sock_fd);
exit(1);
}
printf("server received data from client:%s\n", recvBuffer);
printf("msg.msg_namelen=%d\n", msg.msg_namelen);
printf("new socket fd=%d\n", sock_fd);
strcat(recvBuffer, "--data from server");
if (sendmsg(sock_fd, &msg, 0) == -1) {
perror("sendmsg()\n");
close(sock_fd);
exit(1);
}
}
close(sock_fd);
return 0;
}
Signed-off-by: Weiping Pan <wpan@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
If there is no OVS_SAMPLE_ATTR_ACTIONS set then "acts_list" is NULL and
it leads to a NULL dereference when we call nla_len(acts_list). This
is a static checker fix, not something I have seen in testing.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
ICMP messages generated in output path if frame length is bigger than
mtu are actually lost because socket is owned by user (doing the xmit)
One example is the ipgre_tunnel_xmit() calling
icmp_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED, htonl(mtu));
We had a similar case fixed in commit a34a101e1e6 (ipv6: disable GSO on
sockets hitting dst_allfrag).
Problem of such fix is that it relied on retransmit timers, so short tcp
sessions paid a too big latency increase price.
This patch uses the tcp_release_cb() infrastructure so that MTU
reduction messages (ICMP messages) are not lost, and no extra delay
is added in TCP transmits.
Reported-by: Maciej Żenczykowski <maze@google.com>
Diagnosed-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Nandita Dukkipati <nanditad@google.com>
Cc: Tom Herbert <therbert@google.com>
Cc: Tore Anderson <tore@fud.no>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
The 57840 boards come in two flavours: 2 x 20G and 4 x 10G.
To better differentiate between the two flavours, a separate device ID
was assigned to each.
The silicon default value is still the currently supported 57840 device ID
(0x168d), and since a user can damage the nvram (e.g., 'ethtool -E')
the driver will still support this device ID to allow the user to amend the
nvram back into a supported configuration.
Notice this patch contains lines longer than 80 characters (strings).
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>
|
|
In tcp_tw_remember_stamp we incorrectly checked tw
instead of tm, it can lead to oops if the cached entry is
not found.
tcpm_stamp was not updated in tcpm_check_stamp when
tcpm_suck_dst was called, move the update into tcpm_suck_dst,
so that we do not call it infinitely on every next cache hit
after TCP_METRICS_TIMEOUT.
Signed-off-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
return value
Change niu_rbr_fill() to use unlikely() to check niu_rbr_add_page() return
value to be consistent with the rest of the checks after niu_rbr_add_page()
calls in this file.
Signed-off-by: Shuah Khan <shuah.khan@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Fix Neptune ethernet driver to check dma mapping error after map_page()
interface returns.
Signed-off-by: Shuah Khan <shuah.khan@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
'ocrdma' and 'qib' into for-linus
|
|
Acked-by: Chris Metcalf <cmetcalf@tilera.com>
Signed-off-by: Cong Wang <amwang@redhat.com>
|
|
All callers of kmap_atomic_primary() use __KM_CACHE, so it can be
removed safely, and __kmap_atomic_primary() only check if 'type' if
__KM_CACHE or not, so 'type' can be changed to a boolean as well.
Ditto for kunmap_atomic_primary()/__kunmap_atomic_primary().
Acked-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Cong Wang <amwang@redhat.com>
|
|
Signed-off-by: Cong Wang <amwang@redhat.com>
|
|
The current mpic code creates a linear revmap just big enough for all
the sources, which happens to miss the IPIs and timers on some machines.
This will in turn break when the irqdomain code loses the fallback of
doing a linear search when the revmap fails (and really slows down IPIs
otherwise).
This happens for example on the U4 based Apple machines such as the
dual core PowerMac G5s.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
|
|
In net/compat.c::put_cmsg_compat() we may assign 'data' the address of
either the 'ctv' or 'cts' local variables inside the 'if
(!COMPAT_USE_64BIT_TIME)' branch.
Those variables go out of scope at the end of the 'if' statement, so
when we use 'data' further down in 'copy_to_user(CMSG_COMPAT_DATA(cm),
data, cmlen - sizeof(struct compat_cmsghdr))' there's no telling what
it may be refering to - not good.
Fix the problem by simply giving 'ctv' and 'cts' function scope.
Signed-off-by: Jesper Juhl <jj@chaosbits.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
The ipv4 routing cache is non-deterministic, performance wise, and is
subject to reasonably easy to launch denial of service attacks.
The routing cache works great for well behaved traffic, and the world
was a much friendlier place when the tradeoffs that led to the routing
cache's design were considered.
What it boils down to is that the performance of the routing cache is
a product of the traffic patterns seen by a system rather than being a
product of the contents of the routing tables. The former of which is
controllable by external entitites.
Even for "well behaved" legitimate traffic, high volume sites can see
hit rates in the routing cache of only ~%10.
The general flow of this patch series is that first the routing cache
is removed. We build a completely new rtable entry every lookup
request.
Next we make some simplifications due to the fact that removing the
routing cache causes several members of struct rtable to become no
longer necessary.
Then we need to make some amends such that we can legally cache
pre-constructed routes in the FIB nexthops. Firstly, we need to
invalidate routes which are hit with nexthop exceptions. Secondly we
have to change the semantics of rt->rt_gateway such that zero means
that the destination is on-link and non-zero otherwise.
Now that the preparations are ready, we start caching precomputed
routes in the FIB nexthops. Output and input routes need different
kinds of care when determining if we can legally do such caching or
not. The details are in the commit log messages for those changes.
The patch series then winds down with some more struct rtable
simplifications and other tidy ups that remove unnecessary overhead.
On a SPARC-T3 output route lookups are ~876 cycles. Input route
lookups are ~1169 cycles with rpfilter disabled, and about ~1468
cycles with rpfilter enabled.
These measurements were taken with the kbench_mod test module in the
net_test_tools GIT tree:
git://git.kernel.org/pub/scm/linux/kernel/git/davem/net_test_tools.git
That GIT tree also includes a udpflood tester tool and stresses
route lookups on packet output.
For example, on the same SPARC-T3 system we can run:
time ./udpflood -l 10000000 10.2.2.11
with routing cache:
real 1m21.955s user 0m6.530s sys 1m15.390s
without routing cache:
real 1m31.678s user 0m6.520s sys 1m25.140s
Performance undoubtedly can easily be improved further.
For example fib_table_lookup() performs a lot of excessive
computations with all the masking and shifting, some of it
conditionalized to deal with edge cases.
Also, Eric's no-ref optimization for input route lookups can be
re-instated for the FIB nexthop caching code path. I would be really
pleased if someone would work on that.
In fact anyone suitable motivated can just fire up perf on the loading
of the test net_test_tools benchmark kernel module. I spend much of
my time going:
bash# perf record insmod ./kbench_mod.ko dst=172.30.42.22 src=74.128.0.1 iif=2
bash# perf report
Thanks to helpful feedback from Joe Perches, Eric Dumazet, Ben
Hutchings, and others.
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Commit 9778b696a0188ad3b3524b383953ee73b31b7b68 accidentally added
a .rej file (probably my fault), remove it.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/cjb/mmc
Pull MMC updates from Chris Ball:
"MMC highlights for 3.6:
Core:
- Rename cd-gpio to slot-gpio and extend it to support more slot GPIO
functions, such as write-protect.
- Add a function to get regulators (Vdd and Vccq) for a host.
Drivers:
- sdhci-pxav2, sdhci-pxav3: Add device tree support.
- sdhi: Add device tree support.
- sh_mmcif: Add support for regulators, device tree, slot-gpio.
- tmio: Add regulator support, use slot-gpio."
* tag 'mmc-merge-for-3.6-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/cjb/mmc: (62 commits)
mmc: sdhci-dove: Prepare for common clock framework
mmc: sdhci-dove: Add SDHCI_QUIRK_NO_HISPD_BIT
mmc: omap_hsmmc: ensure probe returns error upon resource failure
mmc: mxs-mmc: Add wp-inverted property
mmc: esdhc: Fix DMA_MASK to not break mx25 DMA access
mmc: core: reset signal voltage on power up
mmc: sd: Fix sd current limit setting
mmc: omap_hsmmc: add clk_prepare and clk_unprepare
mmc: sdhci: When a UHS switch fails, cycle power if regulator is used
mmc: atmel-mci: modify CLKDIV displaying in debugfs
mmc: atmel-mci: fix incorrect setting of host->data to NULL
mmc: sdhci: poll for card even when card is logically unremovable
mmc: sdhci: Introduce new flag SDHCI_USING_RETUNING_TIMER
mmc: sdio: Change pr_warning to pr_warn_ratelimited
mmc: core: Simplify and fix for SD switch processing
mmc: sdhci: restore host settings when card is removed
mmc: sdhci: fix incorrect command used in tuning
mmc: sdhci-pci: CaFe has broken card detection
mmc: sdhci: Report failure reasons for all cases in sdhci_add_host()
mmc: s3cmci: Convert s3cmci driver to gpiolib API
...
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86/mce changes from Ingo Molnar:
"This tree improves the AMD thresholding bank code and includes a
memory fault signal handling fixlet."
* 'x86-mce-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/mce: Fix siginfo_t->si_addr value for non-recoverable memory faults
x86, MCE, AMD: Update copyrights and boilerplate
x86, MCE, AMD: Give proper names to the thresholding banks
x86, MCE, AMD: Make error_count read only
x86, MCE, AMD: Cleanup reading of error_count
x86, MCE, AMD: Print decimal thresholding values
x86, MCE, AMD: Move shared bank to node descriptor
x86, MCE, AMD: Remove local_allocate_... wrapper
x86, MCE, AMD: Remove shared banks sysfs linking
x86, amd_nb: Export model 0x10 and later PCI id
|
|
As mach-dove is moving towards common clock framework prepare
the sdhci driver to grab its clock.
Signed-off-by: Sebastian Hesselbarth <sebastian.hesselbarth@googlemail.com>
Signed-off-by: Chris Ball <cjb@laptop.org>
|
|
The sdio controller on dove doesn't have a bit to indicate
high-speed. With the quirk set it fixes accessing high-speed
sdcards.
Signed-off-by: Sebastian Hesselbarth <sebastian.hesselbarth@googlemail.com>
Signed-off-by: Chris Ball <cjb@laptop.org>
|
|
If platform_get_resource_by_name() fails, driver probe is aborted an
should return an error so the driver is not bound to the device.
However, in the current error path of platform_get_resource_by_name(),
probe returns zero since the return value (ret) is not properly set.
With a zero return value, the driver core assumes probe was successful
and will bind the driver to the device.
Fix this by ensuring that probe returns an error code in this failure
path.
Signed-off-by: Kevin Hilman <khilman@ti.com>
Acked-by: Venkatraman S <svenkatr@ti.com>
Signed-off-by: Chris Ball <cjb@laptop.org>
|
|
The write-protect GPIO is inverted on some boards. Handle such case.
Signed-off-by: Marek Vasut <marex@denx.de>
Signed-off-by: Chris Ball <cjb@laptop.org>
|
