The kernel's math-emu code contains a macro _FP_FROM_INT() which is
used to convert an integer to a raw normalized floating-point value.
It does this basically in three steps:
1. Compute the exponent from the number of leading zero bits.
2. Downshift large fractions to put the MSB in the right position
for normalized fractions.
3. Upshift small fractions to put the MSB in the right position.
There is an boundary error in step 2, causing a fraction with its
MSB exactly one bit above the normalized MSB position to not be
downshifted. This results in a non-normalized raw float, which when
packed becomes a massively inaccurate representation for that input.
The impact of this depends on a number of arch-specific factors,
but it is known to have broken emulation of FXTOD instructions
on UltraSPARC III, which was originally reported as GCC bug 44631
<http://gcc.gnu.org/bugzilla/show_bug.cgi?id=44631>.
Any arch which uses math-emu to emulate conversions from integers to
same-size floats may be affected.
The fix is simple: the exponent comparison used to determine if the
fraction should be downshifted must be "<=" not "<".
I'm sending a kernel module to test this as a reply to this message.
There are also SPARC user-space test cases in the GCC bug entry.
Signed-off-by: Mikael Pettersson <mikpe@it.uu.se> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
sky2_phy_reinit is called by the ethtool helpers sky2_set_settings,
sky2_nway_reset and sky2_set_pauseparam when netif_running.
However, at the end of sky2_phy_init GM_GP_CTRL has GM_GPCR_RX_ENA and
GM_GPCR_TX_ENA cleared. So, doing these commands causes the device to
stop working:
$ ethtool -r eth0
$ ethtool -A eth0 autoneg off
Fix this issue by enabling Rx/Tx after running sky2_phy_init in
sky2_phy_reinit.
Signed-off-by: Brandon Philips <bphilips@suse.de> Tested-by: Brandon Philips <bphilips@suse.de> Cc: stable@kernel.org Tested-by: Mike McCormack <mikem@ring3k.org> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
If the call to phy_connect fails, we will return directly instead of freeing
the previously allocated struct net_device.
Signed-off-by: Florian Fainelli <florian@openwrt.org> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
While chasing a bug report involving a OS/2 server, I noticed the server sets
pSMBr->CountHigh to a incorrect value even in case of normal writes. This
results in 'nbytes' being computed wrongly and triggers a kernel BUG at
mm/filemap.c.
Why the server is setting 'CountHigh' is not clear but only does so after
writing 64k bytes. Though this looks like the server bug, the client side
crash may not be acceptable.
The workaround is to mask off high 16 bits if the number of bytes written as
returned by the server is greater than the bytes requested by the client as
suggested by Jeff Layton.
Reviewed-by: Jeff Layton <jlayton@samba.org> Signed-off-by: Suresh Jayaraman <sjayaraman@suse.de> Signed-off-by: Steve French <sfrench@us.ibm.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This bug appears to be the result of a cut-and-paste mistake from the
NTLMv1 code. The function to generate the MAC key was commented out, but
not the conditional above it. The conditional then ended up causing the
session setup key not to be copied to the buffer unless this was the
first session on the socket, and that made all but the first NTLMv2
session setup fail.
Fix this by removing the conditional and all of the commented clutter
that made it difficult to see.
find_keyring_by_name() can gain access to a keyring that has had its reference
count reduced to zero, and is thus ready to be freed. This then allows the
dead keyring to be brought back into use whilst it is being destroyed.
This problem is that find_keyring_by_name does not confirm that the keyring is
valid before accepting it.
Skipping keyrings that have been reduced to a zero count seems the way to go.
To this end, use atomic_inc_not_zero() to increment the usage count and skip
the candidate keyring if that returns false.
The following script _may_ cause the bug to happen, but there's no guarantee
as the window of opportunity is small:
Clear the floating point exception flag before returning to
user space. This is needed, else the libc trampoline handler
may hit the same SIGFPE again while building up a trampoline
to a signal handler.
What happens is that, when the tipc module in inserted it enters a standalone
node mode in which communication to its own address is allowed <0.0.0> but not
to other addresses, since the appropriate data structures have not been
allocated yet (specifically the tipc_net pointer). There is nothing stopping a
client from trying to send such a message however, and if that happens, we
attempt to dereference tipc_net.zones while the pointer is still NULL, and
explode. The fix is pretty straightforward. Since these oopses all arise from
the dereference of global pointers prior to their assignment to allocated
values, and since these allocations are small (about 2k total), lets convert
these pointers to static arrays of the appropriate size. All the accesses to
these bits consider 0/NULL to be a non match when searching, so all the lookups
still work properly, and there is no longer a chance of a bad dererence
anywhere. As a bonus, this lets us eliminate the setup/teardown routines for
those pointers, and elimnates the need to preform any locking around them to
prevent access while their being allocated/freed.
I've updated the tipc_net structure to behave this way to fix the exact reported
problem, and also fixed up the tipc_bearers and media_list arrays to fix an
obvious simmilar problem that arises from issuing tipc-config commands to
manipulate bearers/links prior to entering networked mode
I've tested this for a few hours by running the sanity tests and stress test
with the tipcutils suite, and nothing has fallen over. There have been a few
lockdep warnings, but those were there before, and can be addressed later, as
they didn't actually result in any deadlock.
Signed-off-by: Neil Horman <nhorman@tuxdriver.com> CC: Allan Stephens <allan.stephens@windriver.com> CC: David S. Miller <davem@davemloft.net> CC: tipc-discussion@lists.sourceforge.net Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Add a new UMOUNT_NOFOLLOW flag to umount(2). This is needed to prevent
symlink attacks in unprivileged unmounts (fuse, samba, ncpfs).
Additionally, return -EINVAL if an unknown flag is used (and specify
an explicitly unused flag: UMOUNT_UNUSED). This makes it possible for
the caller to determine if a flag is supported or not.
CC: Eugene Teo <eugene@redhat.com> CC: Michael Kerrisk <mtk.manpages@gmail.com> Signed-off-by: Miklos Szeredi <mszeredi@suse.cz> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Its an skb_over_panic BUG halt that results from processing an init chunk in
which too many of its variable length parameters are in some way malformed.
The problem is in sctp_process_unk_param:
if (NULL == *errp)
*errp = sctp_make_op_error_space(asoc, chunk,
ntohs(chunk->chunk_hdr->length));
if (*errp) {
sctp_init_cause(*errp, SCTP_ERROR_UNKNOWN_PARAM,
WORD_ROUND(ntohs(param.p->length)));
sctp_addto_chunk(*errp,
WORD_ROUND(ntohs(param.p->length)),
param.v);
When we allocate an error chunk, we assume that the worst case scenario requires
that we have chunk_hdr->length data allocated, which would be correct nominally,
given that we call sctp_addto_chunk for the violating parameter. Unfortunately,
we also, in sctp_init_cause insert a sctp_errhdr_t structure into the error
chunk, so the worst case situation in which all parameters are in violation
requires chunk_hdr->length+(sizeof(sctp_errhdr_t)*param_count) bytes of data.
The result of this error is that a deliberately malformed packet sent to a
listening host can cause a remote DOS, described in CVE-2010-1173:
http://cve.mitre.org/cgi-bin/cvename.cgi?name=2010-1173
I've tested the below fix and confirmed that it fixes the issue. We move to a
strategy whereby we allocate a fixed size error chunk and ignore errors we don't
have space to report. Tested by me successfully
Signed-off-by: Neil Horman <nhorman@tuxdriver.com> Acked-by: Vlad Yasevich <vladislav.yasevich@hp.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
With delayed allocation we lock the page in write_cache_pages() and
try to build an in memory extent of contiguous blocks. This is needed
so that we can get large contiguous blocks request. If range_cyclic
mode is enabled, write_cache_pages() will loop back to the 0 index if
no I/O has been done yet, and try to start writing from the beginning
of the range. That causes an attempt to take the page lock of lower
index page while holding the page lock of higher index page, which can
cause a dead lock with another writeback thread.
The solution is to implement the range_cyclic behavior in
ext4_da_writepages() instead.
http://bugzilla.kernel.org/show_bug.cgi?id=12579
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu> Signed-off-by: Jayson R. King <dev@jaysonking.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
The range_cyclic writeback mode uses the address_space writeback_index
as the start index for writeback. With delayed allocation we were
updating writeback_index wrongly resulting in highly fragmented file.
This patch reduces the number of extents reduced from 4000 to 27 for a
3GB file.
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: Theodore Ts'o <tytso@mit.edu>
[dev@jaysonking.com: Some changed lines from the original version of this patch were dropped, since they were rolled up with another cherry-picked patch applied to 2.6.27.y earlier.]
[dev@jaysonking.com: Use of wbc->no_nrwrite_index_update was dropped, since write_cache_pages_da() implies it.] Signed-off-by: Jayson R. King <dev@jaysonking.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Make a copy of write_cache_pages() for the benefit of
ext4_da_writepages(). This allows us to simplify the code some, and
will allow us to further customize the code in future patches.
There are some nasty hacks in write_cache_pages(), which Linus has
(correctly) characterized as vile. I've just copied it into
write_cache_pages_da(), without trying to clean those bits up lest I
break something in the ext4's delalloc implementation, which is a bit
fragile right now. This will allow Dave Chinner to clean up
write_cache_pages() in mm/page-writeback.c, without worrying about
breaking ext4. Eventually write_cache_pages_da() will go away when I
rewrite ext4's delayed allocation and create a general
ext4_writepages() which is used for all of ext4's writeback. Until
now this is the lowest risk way to clean up the core
write_cache_pages() function.
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu> Cc: Dave Chinner <david@fromorbit.com>
[dev@jaysonking.com: Dropped the hunks which reverted the use of no_nrwrite_index_update, since those lines weren't ever created on 2.6.27.y]
[dev@jaysonking.com: Copied from 2.6.27.y's version of write_cache_pages(), plus the changes to it from patch "vfs: Add no_nrwrite_index_update writeback control flag"] Signed-off-by: Jayson R. King <dev@jaysonking.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
If groups_per_flex < 2, sbi->s_flex_groups[] doesn't get filled out,
and every other access to this first tests s_log_groups_per_flex;
same thing needs to happen in resize or we'll wander off into
a null pointer when doing an online resize of the file system.
Thanks to Christoph Biedl, who came up with the trivial testcase:
Making gconfig fails on fedora 13 as the linker cannot resolve dlsym.
Adding libdl to the link command fixes this.
make shows this error :-
/usr/bin/ld: scripts/kconfig/kconfig_load.o: undefined reference to symbol 'dlsym@@GLIBC_2.2.5'
/usr/bin/ld: note: 'dlsym@@GLIBC_2.2.5' is defined in DSO /lib64/libdl.so.2 so try adding it to the linker command line
/lib64/libdl.so.2: could not read symbols: Invalid operation
tested on x86_64 fedora 13.
Signed-off-by: Richard Kennedy <richard@rsk.demon.co.uk> Reviewed-by: WANG Cong <xiyou.wangcong@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Michal Marek <mmarek@suse.cz> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
I/O errors can happen due to temporary failures, like multipath
errors or losing network contact with the iSCSI server. Because
of that, the VM will retry readpage on the page.
However, do_generic_file_read does not clear PG_error. This
causes the system to be unable to actually use the data in the
page cache page, even if the subsequent readpage completes
successfully!
The function filemap_fault has had a ClearPageError before
readpage forever. This patch simply adds the same to
do_generic_file_read.
Signed-off-by: Jeff Moyer <jmoyer@redhat.com> Signed-off-by: Rik van Riel <riel@redhat.com> Acked-by: Larry Woodman <lwoodman@redhat.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
When the user sets the block device to readwrite then the mddev should
follow suit. Otherwise, the BUG_ON in md_write_start() will be set to
trigger.
The reverse direction, setting mddev->ro to match a set readonly
request, can be ignored because the blkdev level readonly flag precludes
the need to have mddev->ro set correctly. Nevermind the fact that
setting mddev->ro to 1 may fail if the array is in use.
Signed-off-by: Dan Williams <dan.j.williams@intel.com> Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
read_balance uses a "unsigned long" for a sector number which
will get truncated beyond 2TB.
This will cause read-balancing to be non-optimal, and can cause
data to be read from the 'wrong' branch during a resync. This has a
very small chance of returning wrong data.
Reported-by: Jordan Russell <jr-list-2010@quo.to> Signed-off-by: NeilBrown <neilb@suse.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
There is a very small race window when writing to a
RAID1 such that if a device is marked faulty at exactly the wrong
time, the write-in-progress will not be sent to the device,
but the bitmap (if present) will be updated to say that
the write was sent.
Then if the device turned out to still be usable as was re-added
to the array, the bitmap-based-resync would skip resyncing that
block, possibly leading to corruption. This would only be a problem
if no further writes were issued to that area of the device (i.e.
that bitmap chunk).
Currently we always call start-cpu irrespective of if the CPU is
stopped or not. Unfortunatley on POWER7, firmware seems to not like
start-cpu being called when a cpu already been started. This was not
the case on POWER6 and earlier.
This patch checks to see if the CPU is stopped or not via an
query-cpu-stopped-state call, and only calls start-cpu on CPUs which
are stopped.
This fixes a bug with kexec on POWER7 on PHYP where only the primary
thread would make it to the second kernel.
Reported-by: Ankita Garg <ankita@linux.vnet.ibm.com> Signed-off-by: Michael Neuling <mikey@neuling.org> Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Commit 0119536c, which added the assembly version of strncmp to
powerpc, mentions that it adds two instructions to the version from
boot/string.S to allow it to handle len=0. Unfortunately, it doesn't
always return 0 when that is the case. The length is passed in r5, but
the return value is passed back in r3. In certain cases, this will
happen to work. Otherwise it will pass back the address of the first
string as the return value.
This patch lifts the len <= 0 handling code from memcpy to handle that
case.
Reported by: Christian_Sellars@symantec.com Signed-off-by: Jeff Mahoney <jeffm@suse.com> Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
The SH SOHARD ARCNET cards are implemented using generic PLX Technology
PCI<->IOBus bridges. Subvendor and subdevice IDs were not specified,
causing the driver to attach to any such bridge and likely crash the
system by attempting to initialize an unrelated device.
Fix by specifying subvendor and subdevice according to the values found
in the PCI-ID Repository at http://pci-ids.ucw.cz/ .
Signed-off-by: Andreas Bombe <aeb@debian.org> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
The /proc/fs/nfsd/versions file calls nfsd_vers() to check whether
the particular nfsd version is present/available. The problem is
that once I turn off e.g. NFSD-V4 this call returns -1 which is
true from the callers POV which is wrong.
The proposal is to report false in that case.
The bug has existed since 6658d3a7bbfd1768 "[PATCH] knfsd: remove
nfsd_versbits as intermediate storage for desired versions".
Signed-off-by: Pavel Emelyanov <xemul@openvz.org> Acked-by: NeilBrown <neilb@suse.de> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
There are ATAPI devices which raise AN when hit by commands issued by
open(). This leads to infinite loop of AN -> MEDIA_CHANGE uevent ->
udev open() to check media -> AN.
Both ACS and SerialATA standards don't define in which case ATAPI
devices are supposed to raise or not raise AN. They both list media
insertion event as a possible use case for ATAPI ANs but there is no
clear description of what constitutes such events. As such, it seems
a bit too naive to export ANs directly to userland as MEDIA_CHANGE
events without further verification (which should behave similarly to
windows as it apparently is the only thing that some hardware vendors
are testing against).
This patch adds libata.atapi_an module parameter and disables ATAPI AN
by default for now.
Signed-off-by: Tejun Heo <tj@kernel.org> Cc: Kay Sievers <kay.sievers@vrfy.org> Cc: Nick Bowler <nbowler@elliptictech.com> Cc: David Zeuthen <david@fubar.dk> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
The previous patch from Alan Cox ("nfsd: fix vm overcommit crash",
commit 731572d39fcd3498702eda4600db4c43d51e0b26) fixed the problem where
knfsd crashes on exported shmemfs objects and strict overcommit is set.
But the patch forgot supporting the case when CONFIG_SECURITY is
disabled.
This patch copies a part of his fix which is mainly for detecting a bug
earlier.
Acked-by: James Morris <jmorris@namei.org> Signed-off-by: Alan Cox <alan@redhat.com> Signed-off-by: Junjiro R. Okajima <hooanon05@yahoo.co.jp> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Junjiro R. Okajima reported a problem where knfsd crashes if you are
using it to export shmemfs objects and run strict overcommit. In this
situation the current->mm based modifier to the overcommit goes through a
NULL pointer.
We could simply check for NULL and skip the modifier but we've caught
other real bugs in the past from mm being NULL here - cases where we did
need a valid mm set up (eg the exec bug about a year ago).
To preserve the checks and get the logic we want shuffle the checking
around and add a new helper to the vm_ security wrappers
Also fix a current->mm reference in nommu that should use the passed mm
[akpm@linux-foundation.org: coding-style fixes]
[akpm@linux-foundation.org: fix build] Reported-by: Junjiro R. Okajima <hooanon05@yahoo.co.jp> Acked-by: James Morris <jmorris@namei.org> Signed-off-by: Alan Cox <alan@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Experience has shown that the block buffer can only be used for SMBus
(not I2C) block transactions, even though the datasheet doesn't
mention this limitation.
Reported-by: Felix Rubinstein <felixru@gmail.com> Signed-off-by: Jean Delvare <khali@linux-fr.org> Cc: Oleg Ryjkov <oryjkov@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Different motherboards have different PNP declarations for
W83781D/W83782D chips. Some declare the whole range of I/O ports (8
ports), some declare only the useful ports (2 ports at offset 5) and
some declare fancy ranges, for example 4 ports at offset 4. To
properly handle all cases, request all ports individually for probing.
After we have determined that we really have a W83781D or W83782D
chip, the useful port range will be requested again, as a single
block.
I did not see a board which needs this yet, but I know of one for lm78
driver and I'd like to keep the logic of these two drivers in sync.
Signed-off-by: Jean Delvare <khali@linux-fr.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
A race between svc_revisit and svc_delete_xprt can result in
deferred requests holding references on a transport that can never be
recovered because dead transports are not enqueued for subsequent
processing.
Check for XPT_DEAD in revisit to clean up completing deferrals on a dead
transport and sweep a transport's deferred queue to do the same for queued
but unprocessed deferrals.
Signed-off-by: Tom Tucker <tom@opengridcomputing.com> Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Cc: roma1390 <roma1390@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
libata currently doesn't retry if a command fails with AC_ERR_INVALID
assuming that retrying won't get it any further even if retried.
However, a failure may be classified as invalid through hardware
glitch (incorrect reading of the error register or firmware bug) and
there isn't whole lot to gain by not retrying as actually invalid
commands will be failed immediately. Also, commands serving FS IOs
are extremely unlikely to be invalid. Retry FS IOs even if it's
marked invalid.
Transient and incorrect invalid failure was seen while debugging
firmware related issue on Samsung n130 on bko#14314.
http://bugzilla.kernel.org/show_bug.cgi?id=14314
Signed-off-by: Tejun Heo <tj@kernel.org> Reported-by: Johannes Stezenbach <js@sig21.net> Signed-off-by: Jeff Garzik <jgarzik@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Some FSC hardware monitoring chips (Syleus at least) doesn't like
quick writes we typically use to probe for I2C chips. Use a regular
byte read instead for the address they live at (0x73). These are the
only known chips living at this address on PC systems.
For clarity, this fix should not be needed for kernels 2.6.30 and
later, as we started instantiating the hwmon devices explicitly based
on DMI data. Still, this fix is valuable in the following two cases:
* Support for recent FSC chips on older kernels. The DMI-based device
instantiation is more difficult to backport than the device support
itself.
* Case where the DMI-based device instantiation fails, whatever the
reason. We fall back to probing in that case, so it should work.
This fixes kernel bug #15634:
https://bugzilla.kernel.org/show_bug.cgi?id=15634
Signed-off-by: Jean Delvare <khali@linux-fr.org> Acked-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
NFSv4 mounts ignore the rsize and wsize mount options, and always use
the default transfer size for both. This seems to be because all
NFSv4 mounts are now cloned, and the cloning logic doesn't copy the
rsize and wsize settings from the parent nfs_server.
I tested Fedora's 2.6.32.11-99 and it seems to have this problem as
well, so I'm guessing that .33, .32, and perhaps older kernels have
this issue as well.
If dentry found stale happens to be a root of disconnected tree, we
can't d_drop() it; its d_hash is actually part of s_anon and d_drop()
would simply hide it from shrink_dcache_for_umount(), leading to
all sorts of fun, including busy inodes on umount and oopsen after
that.
Bug had been there since at least 2006 (commit c636eb already has it),
so it's definitely -stable fodder.
There is a typo here. We should be testing "*dentry" which was just
assigned instead of "dentry". This could result in dereferencing an
ERR_PTR inside either usbfs_mkdir() or usbfs_create().
Signed-off-by: Dan Carpenter <error27@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
When read_buf is called to move over to the next page in the pagelist
of an NFSv4 request, it sets argp->end to essentially a random
number, certainly not an address within the page which argp->p now
points to. So subsequent calls to READ_BUF will think there is much
more than a page of spare space (the cast to u32 ensures an unsigned
comparison) so we can expect to fall off the end of the second
page.
We never encountered thsi in testing because typically the only
operations which use more than two pages are write-like operations,
which have their own decoding logic. Something like a getattr after a
write may cross a page boundary, but it would be very unusual for it to
cross another boundary after that.
Signed-off-by: J. Bruce Fields <bfields@citi.umich.edu> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Currently clockevents_notify() is called with interrupts enabled at
some places and interrupts disabled at some other places.
This results in a deadlock in this scenario.
cpu A holds clockevents_lock in clockevents_notify() with irqs enabled
cpu B waits for clockevents_lock in clockevents_notify() with irqs disabled
cpu C doing set_mtrr() which will try to rendezvous of all the cpus.
This will result in C and A come to the rendezvous point and waiting
for B. B is stuck forever waiting for the spinlock and thus not
reaching the rendezvous point.
Fix the clockevents code so that clockevents_lock is taken with
interrupts disabled and thus avoid the above deadlock.
Also call lapic_timer_propagate_broadcast() on the destination cpu so
that we avoid calling smp_call_function() in the clockevents notifier
chain.
This issue left us wondering if we need to change the MTRR rendezvous
logic to use stop machine logic (instead of smp_call_function) or add
a check in spinlock debug code to see if there are other spinlocks
which gets taken under both interrupts enabled/disabled conditions.
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com> Cc: "Brown Len" <len.brown@intel.com> Cc: stable@kernel.org
LKML-Reference: <1250544899.2709.210.camel@sbs-t61.sc.intel.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
I got following warning on ia64 box:
In function 'acpi_processor_power_verify':
642: warning: passing argument 2 of 'smp_call_function_single' from
incompatible pointer type
The problem is that the lapic_timer_propagate_broadcast() has 2 versions:
One is real code that modified in the above commit, and the other is NOP
code that used when !ARCH_APICTIMER_STOPS_ON_C3:
So I got warning because of !ARCH_APICTIMER_STOPS_ON_C3.
We really want to do nothing here on !ARCH_APICTIMER_STOPS_ON_C3, so
modify lapic_timer_propagate_broadcast() of real version to use
smp_call_function_single() in it.
Signed-off-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com> Acked-by: Suresh Siddha <suresh.b.siddha@intel.com> Signed-off-by: Len Brown <len.brown@intel.com> Signed-off-by: Thomas Renninger <trenn@suse.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Wang Sheng-Hui [Tue, 13 Apr 2010 13:04:10 +0000 (21:04 +0800)]
trace: Fix inappropriate substraction on tracing_pages_allocated in trace_free_page()
[No matching upstream git commit id as it was fixed differently due to a
rewrite of the tracing code there.]
For normal case, the code in trace_free_page() do once more substraction
on tracing_pages_allocated, but for CONFIG_TRACER_MAX_TRACE it doesn't
take the freed page into account. That's not consistent with
trace_alloc_page(). Well, for there are no message related with this,
so we cannot observe its incorrect state when the kernel doesn't define
"CONFIG_TRACER_MAX_TRACE". If you add some pr_info() as
trace_alloc_page(), you may notice it.
Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Signed-off-by: Wang Sheng-Hui <crosslonelyover@gmail.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
[SCSI] megaraid_sas: fix 64 bit sense pointer truncation
has caused a problem for 32bit programs with 64bit os -
http://bugzilla.kernel.org/show_bug.cgi?id=15001
fix by converting the user space 32bit pointer to a 64 bit one when
needed.
[jejb: fix up some 64 bit warnings] Signed-off-by: Tomas Henzl <thenzl@redhat.com> Cc: Bo Yang <Bo.Yang@lsi.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Ext4 was the only user of range_cont writeback mode and ext4 switched
to a different method. So remove the range_cont mode which is not used
in the kernel.
When we truncate files, the meta-data blocks released are not reused
untill we commit the truncate transaction. That means delayed get_block
request will return ENOSPC even if we have free blocks left. Force a
journal commit and retry block allocation if we get ENOSPC with free
blocks left.
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: Mingming Cao <cmm@us.ibm.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu> Signed-off-by: Jayson R. King <dev@jaysonking.com> Signed-off-by: Theodore Ts'o <tytso@mit.edu> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
During block reservation if we don't have enough blocks left, retry
block reservation with smaller block counts. This makes sure we try
fallocate and DIO with smaller request size and don't fail early. The
delayed allocation reservation cannot try with smaller block count. So
retry block reservation to handle temporary disk full conditions. Also
print free blocks details if we fail block allocation during writepages.
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: Mingming Cao <cmm@us.ibm.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu> Signed-off-by: Jayson R. King <dev@jaysonking.com> Signed-off-by: Theodore Ts'o <tytso@mit.edu> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This patch adds dirty block accounting using percpu_counters. Delayed
allocation block reservation is now done by updating dirty block
counter. In a later patch we switch to non delalloc mode if the
filesystem free blocks is greater than 150% of total filesystem dirty
blocks
With delayed allocation we need to make sure block are reserved before
we attempt to allocate them. Otherwise we get block allocation failure
(ENOSPC) during writepages which cannot be handled. This would mean
silent data loss (We do a printk stating data will be lost). This patch
updates the DIO and fallocate code path to do block reservation before
block allocation. This is needed to make sure parallel DIO and fallocate
request doesn't take block out of delayed reserve space.
When free blocks count go below a threshold we switch to a slow patch
which looks at other CPU's accumulated percpu counter values.
percpu_counter_sum_and_set() and percpu_counter_sum() is the same except
the former updates the global counter after accounting. Since we are
taking the fbc->lock to calculate the precise value of the counter in
percpu_counter_sum() anyway, it should simply set fbc->count too, as the
percpu_counter_sum_and_set() does.
This patch merges these two interfaces into one.
Signed-off-by: Mingming Cao <cmm@us.ibm.com> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: <linux-ext4@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu> Signed-off-by: Jayson R. King <dev@jaysonking.com> Signed-off-by: Theodore Ts'o <tytso@mit.edu>
We are a bit agressive in invalidating all the pages. But
it is ok because we really don't know why the block allocation
failed and it is better to come of the writeback path
so that user can look for more info.
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: Jayson R. King <dev@jaysonking.com> Signed-off-by: Theodore Ts'o <tytso@mit.edu> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
The original code doesn't take into consideration that the value of
MIXART_BA0_SIZE - pos can be less than zero which would lead to a large
unsigned value for "count".
Also I moved the check that read size is a multiple of 4 bytes below
the code that adjusts "count".
Add missing newline to dev_warn() message string. This is more of an issue
with older kernels that don't automatically add a newline if it was missing
from the end of the previous line.
Signed-off-by: Dean Nelson <dnelson@redhat.com> Signed-off-by: Jean Delvare <khali@linux-fr.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This patch (as1352) fixes a bug in the way isochronous input data is
returned to userspace for usbfs transfers. The entire buffer must be
copied, not just the first actual_length bytes, because the individual
packets will be discontiguous if any of them are short.
Reported-by: Markus Rechberger <mrechberger@gmail.com> Signed-off-by: Alan Stern <stern@rowland.harvard.edu> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
commit 71fe804b6d5 (mempolicy: use struct mempolicy pointer in
shmem_sb_info) added mpol=local mount option. but its feature is broken
since it was born. because such code always return 1 (i.e. mount
failure).
Fix an 'oops' when a tmpfs mount point is mounted with the mpol=default
mempolicy.
Upon remounting a tmpfs mount point with 'mpol=default' option, the mount
code crashed with a null pointer dereference. The initial problem report
was on 2.6.27, but the problem exists in mainline 2.6.34-rc as well. On
examining the code, we see that mpol_new returns NULL if default mempolicy
was requested. This 'NULL' mempolicy is accessed to store the node mask
resulting in oops.
About 50% of shutdowns of b44 Ethernet adapter ends by kernel panic
with kernels compiled with stack-protector.
Checking b44_magic_pattern() return values, one call of
b44_magic_pattern() returns 127. It means, that set_bit(128, pmask)
was called on line 1509. It means that bit 0 of 17th byte of pmask was
overwritten. But pmask has only 16 bytes. Stack corruption happens.
It seems that set_bit() on line 1509 always writes one bit off.
The fix does not only solve the stack corruption, but also makes Wake
On LAN working on my onboard B44 on Asus A7V-333X mainboard.
It seems that this problem affects all kernel versions since commit 725ad800 ([PATCH] b44: add wol for old nic) on 2006-06-20.
Signed-off-by: Stanislav Brabec <sbrabec@suse.cz> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
this patch fixes a memory leak which occurs when an em28xx card with DVB
extension is unplugged or its DVB extension driver is unloaded. In
dvb_fini(), dev->dvb must be freed before being set to NULL, as is done
in dvb_init() in case of error.
Note that this bug is also present in the latest stable kernel release.
Quote from Documentation/networking/bonding.txt:
"Note that when a bonding interface has no active links, the
driver will immediately reuse the first link that goes up, even if the
updelay parameter has been specified (the updelay is ignored in this
case). If there are slave interfaces waiting for the updelay timeout
to expire, the interface that first went into that state will be
immediately reused. This reduces down time of the network if the
value of updelay has been overestimated, and since this occurs only in
cases with no connectivity, there is no additional penalty for
ignoring the updelay."
This patch actually changes the behaviour in this way.
Signed-off-by: Jiri Pirko <jpirko@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net> Cc: Jean Delvare <jdelvare@suse.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Modify uid check in do_coredump so as to not apply it in the case of
pipes.
This just got noticed in testing. The end of do_coredump validates the
uid of the inode for the created file against the uid of the crashing
process to ensure that no one can pre-create a core file with different
ownership and grab the information contained in the core when they
shouldn' tbe able to. This causes failures when using pipes for a core
dumps if the crashing process is not root, which is the uid of the pipe
when it is created.
The fix is simple. Since the check for matching uid's isn't relevant for
pipes (a process can't create a pipe that the uermodehelper code will open
anyway), we can just just skip it in the event ispipe is non-zero
Reverts a pipe-affecting change which was accidentally made in
Do not set current->mm->mmap to NULL in 32-bit emulation on 64-bit
load_aout_binary after flush_old_exec as it would destroy already
set brpm mapping with arguments.
Matt T. Yourst notes that kvm_arch_vcpu_ioctl_set_sregs lacks validity
checking for the new cr3 value:
"Userspace callers of KVM_SET_SREGS can pass a bogus value of cr3 to
the kernel. This will trigger a NULL pointer access in gfn_to_rmap()
when userspace next tries to call KVM_RUN on the affected VCPU and kvm
attempts to activate the new non-existent page table root.
This happens since kvm only validates that cr3 points to a valid guest
physical memory page when code *inside* the guest sets cr3. However, kvm
currently trusts the userspace caller (e.g. QEMU) on the host machine to
always supply a valid page table root, rather than properly validating
it along with the rest of the reloaded guest state."
loff_t is a signed type. If userspace passes a negative ppos, the "count"
range check is weakened. "count"s bigger than HPEE_MAX_LENGTH will pass the check.
Also, if ppos is negative, the readb(eisa_eeprom_addr + *ppos) will poke in random
memory.
Three bytes of uninitialized kernel memory are currently leaked to user
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Reviewed-by: Jiri Pirko <jpirko@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Almost all r128's private ioctls require that the CCE state has
already been initialised. However, most do not test that this has
been done, and will proceed to dereference a null pointer. This may
result in a security vulnerability, since some ioctls are
unprivileged.
This adds a macro for the common initialisation test and changes all
ioctl implementations that require prior initialisation to use that
macro.
Also, r128_do_init_cce() does not test that the CCE state has not
been initialised already. Repeated initialisation may lead to a crash
or resource leak. This adds that test.
Signed-off-by: Ben Hutchings <ben@decadent.org.uk> Signed-off-by: Dave Airlie <airlied@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Debug registers may only be accessed from cpl 0. Unfortunately, vmx will
code to emulate the instruction even though it was issued from guest
userspace, possibly leading to an unexpected trap later.
While we are never normally passed an instruction that exceeds 15 bytes,
smp games can cause us to attempt to interpret one, which will cause
large latencies in non-preempt hosts.
Signed-off-by: Avi Kivity <avi@redhat.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This patch (as1274) simplifies the counting of transaction-error
retries. Now we will count up from 0 to QH_XACTERR_MAX instead of
down from QH_XACTERR_MAX to 0.
The patch also fixes a small bug: qh->xacterr was not getting
initialized for interrupt endpoints.
I notice that the processcompl_compat() function seems to be leaking the
'struct async *as' in the error paths.
I think that the calling convention is fundamentally buggered. The
caller is the one that did the "reap_as()" to get the as thing, the
caller should be the one to free it too.
Freeing it in the caller also means that it very clearly always gets
freed, and avoids the need for any "free in the error case too".
From: Linus Torvalds <torvalds@linux-foundation.org> Cc: Alan Stern <stern@rowland.harvard.edu> Cc: Marcus Meissner <meissner@suse.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
When controlling an industrial radio modem it can be necessary to
manipulate the handshake lines in order to control the radio modem's
transmitter, from userspace.
The transmitter should not be turned off before all characters have been
transmitted. serial8250_tx_empty() was reporting that all characters were
transmitted before they actually were.
===
Discovered in parallel with more testing and analysis by Kees Schoenmakers
as follows:
I ran into an NetMos 9835 serial pci board which behaves a little
different than the standard. This type of expansion board is very common.
"Standard" 8250 compatible devices clear the 'UART_LST_TEMT" bit together
with the "UART_LSR_THRE" bit when writing data to the device.
The NetMos device does it slightly different
I believe that the TEMT bit is coupled to the shift register. The problem
is that after writing data to the device and very quickly after that one
does call serial8250_tx_empty, it returns the wrong information.
My patch makes the test more robust (and solves the problem) and it does
not affect the already correct devices.
Alan:
We may yet need to quirk this but now we know which chips we have a
way to do that should we find this breaks some other 8250 clone with
dodgy THRE.
Signed-off-by: Dick Hollenbeck <dick@softplc.com> Signed-off-by: Alan Cox <alan@linux.intel.com> Cc: Kees Schoenmakers <k.schoenmakers@sigmae.nl> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
dev_dbg outputs dev_name, which is released with device_unregister. This bug
resulted in output like this:
i2c Xy2�0: adapter [SMBus I801 adapter at 1880] unregistered
The right output would be:
i2c i2c-0: adapter [SMBus I801 adapter at 1880] unregistered
Signed-off-by: Thadeu Lima de Souza Cascardo <cascardo@holoscopio.com> Signed-off-by: Jean Delvare <khali@linux-fr.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
While running 20 parallel instances of dd as follows:
#!/bin/bash
for i in `seq 1 20`; do
dd if=/dev/zero of=/export/hda3/dd_$i bs=1073741824 count=1 &
done
wait
on a 16G machine, we noticed that rather than just killing the processes,
the entire kernel went down. Stracing dd reveals that it first does an
mmap2, which makes 1GB worth of zero page mappings. Then it performs a
read on those pages from /dev/zero, and finally it performs a write.
The machine died during the reads. Looking at the code, it was noticed
that /dev/zero's read operation had been changed by 557ed1fa2620dc119adb86b34c614e152a629a80 ("remove ZERO_PAGE") from giving
zero page mappings to actually zeroing the page.
The zeroing of the pages causes physical pages to be allocated to the
process. But, when the process exhausts all the memory that it can, the
kernel cannot kill it, as it is still in the kernel mode allocating more
memory. Consequently, the kernel eventually crashes.
To fix this, I propose that when a fatal signal is pending during
/dev/zero read operation, we simply return and let the user process die.
Signed-off-by: Salman Qazi <sqazi@google.com> Cc: Nick Piggin <nickpiggin@yahoo.com.au> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
[ Modified error return and comment trivially. - Linus] Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Lin Ming reported a 10% OLTP regression against 2.6.27-rc4.
The difference seems to come from different preemption agressiveness,
which affects the cache footprint of the workload and its effective
cache trashing.
Aggresively preempt a task if its avg overlap is very small, this should
avoid the task going to sleep and find it still running when we schedule
back to it - saving a wakeup.
Reported-by: Lin Ming <ming.m.lin@intel.com> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
The SKY2_HW_RAM_BUFFER bit in hw->flags was checked in sky2_mac_init(),
before being set later in sky2_up().
Setting SKY2_HW_RAM_BUFFER in sky2_init() where other hw->flags are set
should avoid this problem recurring.
Signed-off-by: Mike McCormack <mikem@ring3k.org> Acked-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Just like ip_fast_csum, the assembly snippet in csum_ipv6_magic needs a
memory clobber, as it is only passed the address of the buffer, not a
memory reference to the buffer itself.
This caused failures in Hurd's pfinetv4 when we tried to compile it with
gcc-4.3 (bogus checksums).
Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org> Cc: Ingo Molnar <mingo@elte.hu> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: "H. Peter Anvin" <hpa@zytor.com> Acked-by: "David S. Miller" <davem@davemloft.net> Cc: Andi Kleen <andi@firstfloor.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
This appeared after I enabled CONFIG_TCP_MD5SIG and played with it a
bit, so I looked at what might have caused it.
One thing that struck me as strange is tcp_twsk_destructor(), as it
calls tcp_put_md5sig_pool() -- which entails a put_cpu(), causing the
detected imbalance. Found on 2.6.23.9, but 2.6.31 is affected as well,
as far as I can tell.
Signed-off-by: Robert Varga <nite@hq.alert.sk> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
When requesting all prl entries (kprl.addr == INADDR_ANY) and there are
more prl entries than there is space passed from userspace, the existing
code would always copy cmax+1 entries, which is more than can be handled.
This patch makes the kernel copy only exactly cmax entries.
Signed-off-by: Sascha Hlusiak <contact@saschahlusiak.de> Acked-By: Fred L. Templin <Fred.L.Templin@boeing.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
"..., when one process calls sendmsg once to send 43804 bytes of
data and one file descriptor, and another process then calls recvmsg
three times to receive the 16032+16032+11740 bytes, each of those
recvmsg calls returns the file descriptor in the ancillary data. I
confirmed this with strace. The behaviour differs from Linux
2.6.26, where reportedly only one of those recvmsg calls (I think
the first one) returned the file descriptor."
This bug was introduced by a patch from me titled "net: unix: fix inflight
counting bug in garbage collector", commit 6209344f5.
And the reason is, quoting Kalle:
"Before your patch, unix_attach_fds() would set scm->fp = NULL, so
that if the loop in unix_stream_sendmsg() ran multiple iterations,
it could not call unix_attach_fds() again. But now,
unix_attach_fds() leaves scm->fp unchanged, and I think this causes
it to be called multiple times and duplicate the same file
descriptors to each struct sk_buff."
Fix this by introducing a flag that is cleared at the start and set
when the fds attached to the first buffer. The resulting code should
work equivalently to the one on 2.6.26.
Reported-by: Kalle Olavi Niemitalo <kon@iki.fi> Signed-off-by: Miklos Szeredi <mszeredi@suse.cz> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
In ax25_make_new, if kmemdup of digipeat returns an error, there would
be an oops in sk_free while calling sk_destruct, because sk_protinfo
is NULL at the moment; move sk->sk_destruct initialization after this.
BTW of reported-by: Bernard Pidoux F6BVP <f6bvp@free.fr>
Signed-off-by: Jarek Poplawski <jarkao2@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Alan Stern [Mon, 8 Feb 2010 14:43:22 +0000 (09:43 -0500)]
EHCI: fix bug in keeping track of resuming ports
This patch fixes a bug caused by backporting commit cec3a53c7fe794237b582e8e77fc0e48465e65ee (USB: EHCI & UHCI: fix race
between root-hub suspend and port resume) to 2.6.27.stable without
also backporting commit eafe5b99f2135488b21cf17a262c54997c44f784 (USB:
EHCI: fix remote-wakeup support for ARC/TDI core). This extracts the
necessary changes from the earlier patch and backports them.
The symptom of the bug is that the system will fail to suspend more
than once. The problem is caused by setting ehci->reset_done[i] but
never clearing it. When ehci_bus_suspend() sees a nonzero value
there, it assumes this means the port is in the middle of resuming so
it aborts the bus suspend.
In r8169 driver MTU is used to calculate receive buffer size.
Receive buffer size is used to configure hardware incoming packet filter.
For jumbo frames:
Receive buffer size = Max frame size = MTU + 14 (ethernet header) + 4
(vlan header) + 4 (ethernet checksum) = MTU + 22
Bug:
driver for all MTU up to 1536 use receive buffer size 1536
As you can see from formula, this mean all IP packets > 1536 - 22
(for vlan tagged, 1536 - 18 for not tagged) are dropped by hardware
filter.
Example:
host_good> ifconfig eth0 mtu 1536
host_r8169> ifconfig eth0 mtu 1536
host_good> ping host_r8169
Ok
host_good> ping -s 1500 host_r8169
Fail
host_good> ifconfig eth0 mtu 7000
host_r8169> ifconfig eth0 mtu 7000
host_good> ping -s 1500 host_r8169
Ok
Bonus: got rid of magic number 8
Signed-off-by: Raimonds Cicans <ray@apollo.lv> Signed-off-by: David S. Miller <davem@davemloft.net> Cc: Jean Delvare <khali@linux-fr.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
We want to be sure that compiler fetches the limit variable only
once, so add helpers for fetching current and maximal resource
limits which do that.
Add them to sched.h (instead of resource.h) due to circular dependency
sched.h->resource.h->task_struct
Alternative would be to create a separate res_access.h or similar.