14 Feb, 2013
40 commits
-
The s390 architecture is unique in respect to dirty page detection,
it uses the change bit in the per-page storage key to track page
modifications. All other architectures track dirty bits by means
of page table entries. This property of s390 has caused numerous
problems in the past, e.g. see git commit ef5d437f71afdf4a
"mm: fix XFS oops due to dirty pages without buffers on s390".To avoid future issues in regard to per-page dirty bits convert
s390 to a fault based software dirty bit detection mechanism. All
user page table entries which are marked as clean will be hardware
read-only, even if the pte is supposed to be writable. A write by
the user process will trigger a protection fault which will cause
the user pte to be marked as dirty and the hardware read-only bit
is removed.With this change the dirty bit in the storage key is irrelevant
for Linux as a host, but the storage key is still required for
KVM guests. The effect is that page_test_and_clear_dirty and the
related code can be removed. The referenced bit in the storage
key is still used by the page_test_and_clear_young primitive to
provide page age information.For page cache pages of mappings with mapping_cap_account_dirty
there will not be any change in behavior as the dirty bit tracking
already uses read-only ptes to control the amount of dirty pages.
Only for swap cache pages and pages of mappings without
mapping_cap_account_dirty there can be additional protection faults.
To avoid an excessive number of additional faults the mk_pte
primitive checks for PageDirty if the pgprot value allows for writes
and pre-dirties the pte. That avoids all additional faults for
tmpfs and shmem pages until these pages are added to the swap cache.Signed-off-by: Martin Schwidefsky
-
Right now the page table upgrade does not happen if the end address
of a fixed mapping is greater than TASK_SIZE.
Enhance s390_mmap_check() to handle MAP_FIXED mappings correctly.Signed-off-by: Hendrik Brueckner
Reviewed-by: Heiko Carstens
Signed-off-by: Martin Schwidefsky -
Discard exit.data section at run time, not link time, since exit.text
references exit.data and causes this build error:`.exit.data' referenced in section `.exit.text' of drivers/built-in.o:
defined in discarded section `.exit.data' of drivers/built-in.oSigned-off-by: Heiko Carstens
Signed-off-by: Martin Schwidefsky -
Texas Instruments WL1273 I2C FM Radio (RADIO_WL1273) selects
MFD_CORE, which itself depends on GENERIC_HARDIRQS.
So add the dependency to the TI driver as well.Signed-off-by: Heiko Carstens
Acked-by: Mauro Carvalho Chehab
Signed-off-by: Martin Schwidefsky -
s390 version of 855ddb56 "x86: bpf_jit_comp: add vlan tag support".
Signed-off-by: Heiko Carstens
Signed-off-by: Martin Schwidefsky -
The AT91RM9200 driver call devm_request_irq() and therefore should
depend on GENERIC_HARDIRQS to prevent link/compile errors on plaforms
without GENERIC_HARDIRQS.Signed-off-by: Heiko Carstens
Acked-by: David S. Miller
Signed-off-by: Martin Schwidefsky -
The iucv base layer is initialized during the registration of the
first iucv handler. If no handler is registered and the
iucv_reboot_event() notifier is called, a missing check can cause
a kernel panic in iucv_block_cpu(). To solve this issue, check the
IRQ masks invoke iucv_block_cpu() for enabled CPUs only.Signed-off-by: Hendrik Brueckner
Signed-off-by: Martin Schwidefsky -
Just like on other architectures. The intention is that this will reduce
merge conflicts if new config options get added in sorted order as well.Signed-off-by: Heiko Carstens
Signed-off-by: Martin Schwidefsky -
Signed-off-by: Heiko Carstens
Signed-off-by: Martin Schwidefsky -
Signed-off-by: Heiko Carstens
Signed-off-by: Martin Schwidefsky -
When the DASD devices are detached from the driver, then the
dasd_generic_remove function is called. One of the things this
function should do is to remove the DASD specific sysfs attributes,
but this is not done in all cases. This is likely to cause an oops
when at a later point sysfs stumbles over the stale pointers. In
particular this happens when when the modules are unloaded and loaded
again.Signed-off-by: Stefan Weinhuber
Signed-off-by: Martin Schwidefsky -
Loading the pci hotplug module when no devices are present will fail
but unfortunately some hotplug callbacks stay registered to the pci
bus level. Fix this by not letting module loading fail when no pci
devices are present and provide proper {de}registration functions
for these callbacks.Reviewed-by: Gerald Schaefer
Signed-off-by: Sebastian Ott
Signed-off-by: Martin Schwidefsky -
Use the __get_free_pages wrapper in clp_alloc_block. Also change the
allocation to use one page only. This page is used as CLP response
block e.g. to list available pci functions. Using one page we can
list > 250 pci functions at once and we have code to loop around this
CLP command (if not all functions fit into to the CLP block) already
in place.Acked-by: Gerald Schaefer
Signed-off-by: Sebastian Ott
Signed-off-by: Martin Schwidefsky -
Tell gcc that the memory region pointed to by req will be used (and
changed). Also remove the (now) superfluous memory constraint.Acked-by: Gerald Schaefer
Signed-off-by: Sebastian Ott
Signed-off-by: Martin Schwidefsky -
The CPU-measurement counter facility does not support sampling events
and returns -EINVAL in that case. This return code lets the perf tool
fail. To fall back to software sampling events, return -ENOENT instead.Signed-off-by: Hendrik Brueckner
Signed-off-by: Martin Schwidefsky -
Only needed to make some drivers compile...
Signed-off-by: Heiko Carstens
Signed-off-by: Martin Schwidefsky -
Signed-off-by: Heiko Carstens
Signed-off-by: Martin Schwidefsky -
Add s390 to the list of architectures that don' want ATA.
Signed-off-by: Heiko Carstens
Signed-off-by: Martin Schwidefsky -
pci_probe is too generic and has a name clash with other common code parts.
Signed-off-by: Heiko Carstens
Signed-off-by: Martin Schwidefsky -
There is no such function nor any caller in the whole kernel.
Signed-off-by: Heiko Carstens
Signed-off-by: Martin Schwidefsky -
Provide empty dma_cache_sync() function.
Acked-by: Sebastian Ott
Signed-off-by: Heiko Carstens
Signed-off-by: Martin Schwidefsky -
Some of the now available common code drivers only compile if mb() is a define.
Signed-off-by: Heiko Carstens
Signed-off-by: Martin Schwidefsky -
Fix name clash with some common code device drivers and add "tod"
to all tod clock access function names.Signed-off-by: Heiko Carstens
Signed-off-by: Martin Schwidefsky -
Convert the defines for the !CONFIG_SCM* stuff to static inline
functions.Signed-off-by: Sebastian Ott
Signed-off-by: Martin Schwidefsky -
When an attempt to query or modify the grouping state of a channel path
fails due to a timeout, we cannot be sure about its state. To get back to
a defined state, disband the whole path group and try again while excluding
the offending path.Reviewed-by: Peter Oberparleiter
Signed-off-by: Sebastian Ott
Signed-off-by: Martin Schwidefsky -
Add new attribute "vpm" to the subchannel sysfs directory of I/O
subchannels. This attribute contains a path mask indicating which
channel paths were successfully verified to be usable for I/O.Reviewed-by: Peter Oberparleiter
Signed-off-by: Sebastian Ott
Signed-off-by: Martin Schwidefsky -
Omit known to be broken paths (those set in path_noirq_mask) for the
sense/set PGID and nop IO commands.Note: path_noirq_mask will be reset in ccw_device_verify_start (the
paths could be healthy again). However if we restart a path
verification via verify_start this mask will not be reset (there is
no need to let the wait for an interrupt time out again - plus we
do not want to loop once we deal with the paths in unknown path group
state).Reviewed-by: Peter Oberparleiter
Signed-off-by: Sebastian Ott
Signed-off-by: Martin Schwidefsky -
Do not abort path verification when waiting for an interrupt timed out.
Use path_noirq_mask to keep track of the paths used for this (also
maintain a path_notoper_mask for debugging purposes). If the timeout
happend to be during an operation where we query or alter the state of
path groups set the pgid_unknown flag.With this change we allow usage of devices which have such ill-behaved
paths (if at least one path is operational).Reviewed-by: Peter Oberparleiter
Signed-off-by: Sebastian Ott
Signed-off-by: Martin Schwidefsky -
Under LPAR the zfcpdump HSA is a shared resource. Up to now the HSA memory
is released when the zcore file is closed. Dump programs that know that
they do not need the HSA memory any more (e.g. because they already dumped it)
could release it earlier. This would allow other LPARs to use it again.To achieve this a new debugfs file "hsa" is added that can be used to read
the HSA size and to release the HSA by writing "0" into the file.Signed-off-by: Michael Holzheu
Signed-off-by: Martin Schwidefsky -
When a zfcpdump is triggered and a second dump on the same CEC is
already in progress for another LPAR, diagnose 308 returns with
an error code until the first dump is finished. Currently the
second Linux stops with a disabled wait PSW in that case.This is improved now by by triggering diag 308 in a loop until
it works.Signed-off-by: Michael Holzheu
Signed-off-by: Martin Schwidefsky -
Cleanup the functions used to call SEI.
Also provide !CONFIG_PCI dummys for pci error handling.Reviewed-by: Peter Oberparleiter
Signed-off-by: Sebastian Ott
Signed-off-by: Martin Schwidefsky -
Signed-off-by: Ingo Tuchscherer
Signed-off-by: Martin Schwidefsky -
Given enough debug options some modules can grow large enough
that the GOT table gets bigger than 4K. On s390 the modules
are compiled with -fpic which limits the GOT to 4K. The end
result is a module that is loaded but won't work.Add a sanity check to apply_rela and return with an error if
a relocation error is detected for a module.Signed-off-by: Martin Schwidefsky
-
Convert the synchronous size sense code to an interrupt driven
approach. This allows to set the device online even if the
terminal is not connected. With the new code views can be
registered without a connected terminal, the tty can be opened
as soon as the device is online. After the terminal has been
connected and the size has been determined the tty is resized
to match the device characteristics..Signed-off-by: Martin Schwidefsky
-
Add a notifier to create / destroy the device nodes for the tty view
and the fullscreen view. Only device nodes for online devices are
created and the device names will follow the convention as outlined
in Documentation/devices.txt: 3270/tty for the tty nodes,
3270/tub for hte fullscreen nodes and 3270/tub for the fullscreen
control node.Signed-off-by: Martin Schwidefsky
-
Corrects the order of tasklet_init vs. the allocation of the
read request which has been broken by git commit 9d2ae233
"TTY: tty3270, move initialization to allocation".Signed-off-by: Martin Schwidefsky
-
Reintroduce the tty3270_open function which has been removed by
git commit 20cda6f2 "TTY: tty3270, add tty install". Without
the open function in the tty_operations tty_open will return
-ENODEV and the 3270 tty will not work.Signed-off-by: Martin Schwidefsky
-
E.g. readl is defined like this
#define readl(addr) __le32_to_cpu(__raw_readl(addr))
If a there is a readl() call that doesn't check the return value
this will cause a compile warning on big endian machines due to
the __le32_to_cpu macro magic.E.g. code like this:
readl(addr);
will generate the following compile warning:
warning: value computed is not used [-Wunused-value]
With this patch we get rid of dozens of compile warnings on s390.
Signed-off-by: Heiko Carstens
Acked-by: Arnd Bergmann
Signed-off-by: Martin Schwidefsky -
Pull networking fixes from David Miller:
"This is primarily to get those r8169 reverts sorted, but other fixes
have accumulated meanwhile.1) Revert two r8169 changes to fix suspend/resume for some users,
from Francois Romieu.2) PCI dma mapping errors in atl1c are not checked for and this cause
hard crashes for some users, from Xiong Huang.3) In 3.8.x we merged the removal of the EXPERIMENTAL dependency for
'dlm' but the same patch for 'sctp' got lost somewhere, resulting
in the potential for build errors since there are cross
dependencies. From Kees Cook.4) SCTP's ipv6 socket route validation makes boolean tests
incorrectly, fix from Daniel Borkmann.5) mac80211 does sizeof(ptr) instead of (sizeof(ptr) * nelem), from
Cong Ding.6) arp_rcv() can crash on shared non-linear packets, from Eric
Dumazet.7) Avoid crashes in macvtap by setting ->gso_type consistently in
ixgbe, qlcnic, and bnx2x drivers. From Michael S Tsirkin and
Alexander Duyck.8) Trinity fuzzer spots infinite loop in __skb_recv_datagram(), fix
from Eric Dumazet.9) STP protocol frames should use high packet priority, otherwise an
overloaded bridge can get stuck. From Stephen Hemminger.10) The HTB packet scheduler was converted some time ago to store
internal timestamps in nanoseconds, but we don't convert back into
psched ticks for the user during dumps. Fix from Jiri Pirko.11) mwl8k channel table doesn't set the .band field properly,
resulting in NULL pointer derefs. Fix from Jonas Gorski.12) mac80211 doesn't accumulate channels properly during a scan so we
can downgrade heavily to a much less desirable connection speed.
Fix from Johannes Berg.13) PHY probe failure in stmmac can result in resource leaks and
double MDIO registery later, from Giuseppe CAVALLARO.14) Correct ipv6 checksumming in ip6t_NPT netfilter module, also fix
address prefix mangling, from YOSHIFUJI Hideaki."* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (27 commits)
net, sctp: remove CONFIG_EXPERIMENTAL
net: sctp: sctp_v6_get_dst: fix boolean test in dst cache
batman-adv: Fix NULL pointer dereference in DAT hash collision avoidance
net/macb: fix race with RX interrupt while doing NAPI
atl1c: add error checking for pci_map_single functions
htb: fix values in opt dump
ixgbe: Only set gso_type to SKB_GSO_TCPV4 as RSC does not support IPv6
net: fix infinite loop in __skb_recv_datagram()
net: qmi_wwan: add Yota / Megafon M100-1 4g modem
mwl8k: fix band for supported channels
bridge: set priority of STP packets
mac80211: fix channel selection bug
arp: fix possible crash in arp_rcv()
bnx2x: set gso_type
qlcnic: set gso_type
ixgbe: fix gso type
stmmac: mdio register has to fail if the phy is not found
stmmac: fix macro used for debugging the xmit
Revert "r8169: enable internal ASPM and clock request settings".
Revert "r8169: enable ALDPS for power saving".
... -
Pull x86 fixes from Peter Anvin:
"One (hopefully) last batch of x86 fixes. You asked for the patch by
patch justifications, so here they are:x86, MCE: Retract most UAPI exports
This one unexports from userspace a bunch of definitions which should
never have been exported. We really don't want to create an
accidental legacy here.x86, doc: Add a bootloader ID for OVMF
This is a documentation-only patch, just recording the official
assignment of a boot loader ID.x86: Do not leak kernel page mapping locations
Security: avoid making it needlessly easy for user space to probe the
kernel memory layout.x86/mm: Check if PUD is large when validating a kernel address
Prevent failures using /proc/kcore when using 1G pages.
x86/apic: Work around boot failure on HP ProLiant DL980 G7 Server systems
Works around a BIOS problem causing boot failures on affected hardware."
* 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/mm: Check if PUD is large when validating a kernel address
x86/apic: Work around boot failure on HP ProLiant DL980 G7 Server systems
x86, doc: Add a bootloader ID for OVMF
x86: Do not leak kernel page mapping locations
x86, MCE: Retract most UAPI exports
