The sequence to move over to the Linux trap tables from
the firmware ones needs to be more air tight. It turns
out that to be %100 safe we do need to be able to translate
OBP mappings in our TLB miss handlers early.
In order not to eat up a lot of kernel image memory with
static page tables, just use the translations array in
the OBP TLB miss handlers. That solves the bulk of the
problem.
Furthermore, to make sure the OBP TLB miss path will work
even before the fixed MMU globals are loaded, explicitly
load %g1 to TLB_SFSR at the beginning of the i-TLB and
d-TLB miss handlers.
To ease the OBP TLB miss walking of the prom_trans[] array,
we sort it then delete all of the non-OBP entries in there
(for example, there are entries for the kernel image itself
which we're not interested in at all).
We also save about 32K of kernel image size with this change.
Not a bad side effect :-)
There are still some reasons why trampoline.S can't use the
setup_trap_table() yet. The most noteworthy are:
1) OBP boots secondary processors with non-bias'd stack for
some reason. This is easily fixed by using a small bootup
stack in the kernel image explicitly for this purpose.
2) Doing a firmware call via the normal C call prom_set_trap_table()
goes through the whole OBP enter/exit sequence that saves and
restores OBP and Linux kernel state in the MMUs. This path
unfortunately does a "flush %g6" while loading up the OBP locked
TLB entries for the firmware call.
If we setup the %g6 in the trampoline.S code properly, that
is in the PAGE_OFFSET linear mapping, but we're not on the
kernel trap table yet so those addresses won't translate properly.
One idea is to do a by-hand firmware call like we do in the
early bootup code and elsewhere here in trampoline.S But this
fails as well, as aparently the secondary processors are not
booted with OBP's special locked TLB entries loaded. These
are necessary for the firwmare to processes TLB misses correctly
up until the point where we take over the trap table.
This does need to be resolved at some point.
Signed-off-by: David S. Miller <davem@davemloft.net>
Glibc is about to get some new high precision timer stuff that relies on
the standard timebase of the PPC architecture.
However, some (rare & old) CPUs do not have such timebase and it is a
bit annoying to have your stuff just crash because you are running on
the wrong CPU...
This exposes to userland a CPU feature bit that tells that the current
processor doesn't have a standard timebase. It's negative logic so that
glibc will still "just work" on older kernels (it will just be unhappy
on those old CPUs but that doesn't really matter as distro tend to
update glibc & kernel at the same time).
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Interestingly enough, ppc32 had broken timekeeping for ages... It
worked, but probably drifted a bit more than could be explained by the
actual bad precision of the timebase calibration. We discovered that
recently when somebody figured out that the common code was using
CLOCK_TICK_RATE to correct the timekeeing, and ppc32 had a completely
bogus value for it.
This patch turns it into something saner. Probably not as good as doing
something based on the actual timebase frequency precision but I'll
leave that sort of math to others. This at least makes it better for
the common HZ values.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Give an empty definition for clear_can_do_skas() when it is not needed.
Thanks to Junichi Uekawa <dancer@netfort.gr.jp> for reporting the
breakage and providing a fix (I re-fixed it in an IMHO cleaner way).
Signed-off-by: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The patch to use host AIO support that I submitted early after 2.6.13 exposed
some problems in the block driver. I have fixes for these, but am not
comfortable putting them into 2.6.14 at this late date. So, this patch reverts
the use of host AIO.
I will resubmit the original patch, plus fixes to the driver after 2.6.14
in order to get a reasonable amount of testing before they're exposed to
the general public.
Signed-off-by: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
We were not doing alignment properly when remapping the kernel image.
What we want is a 4MB aligned physical address to map at KERNBASE.
Mistakedly we were 4MB aligning the virtual address where the kernel
initially sits, that's wrong.
Instead, we should PAGE align the virtual address, then 4MB align the
physical address result the prom gives to us.
Signed-off-by: David S. Miller <davem@davemloft.net>
Refuse to install a page into a mapping if the mapping count is already
ridiculously large.
You probably cannot trigger this on 32-bit architectures, but on a
64-bit setup we should protect against it.
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Newer gcc's are generating this relocation, so the module loader needs to
handle it.
Signed-off-by: Peter Bergner <bergner@vnet.ibm.com>
Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Nir Tzachar <tzachar@cs.bgu.ac.il> points out that if an ELF file specifies a
zero-length bss at a whacky address, we cannot load that binary because
padzero() tries to zero out the end of the page at the whacky address, and
that may not be writeable.
See also http://bugzilla.kernel.org/show_bug.cgi?id=5411
So teach load_elf_binary() to skip the bss settng altogether if the elf file
has a zero-length bss segment.
Cc: Roland McGrath <roland@redhat.com>
Cc: Daniel Jacobowitz <dan@debian.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
I've noticed that the calculations for seg_size and nr_segs in
__dma_sync_page_highmem() (arch/ppc/kernel/dma-mapping.c) are wrong. The
incorrect calculations can result in either an oops or a panic when running
fsck depending on the size of the partition.
The problem with the seg_size calculation is that it can result in a
negative number if size is offset > size. The problem with the nr_segs
caculation is returns the wrong number of segments, e.g. it returns 1 when
size is 200 and offset is 4095, when it should return 2 or more.
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Revert this recent correctness change: Douglas Crosher <dcrosher@scieneer.com>
reported that it broke an existing application, and that madvise() works
without error on anonymous mappings on Solaris.
This means that madvise() will remain non-standards-compliant: we should
return -EBADF for all requests against non-file-backed vma's, but Linux only
does this for MADV_WILLNEED requests.
Signed-off-by: Suzuki K P <suzuki@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Here is a compatibility fix between Linux and Solaris when used with VxFS
filesystems: Solaris usually accepts acl entries in any order, but with
VxFS it replies with NFSERR_INVAL when it sees a four-entry acl that is not
in canonical form. It may also fail with other non-canonical acls -- I
can't tell, because that case never triggers: We only send non-canonical
acls when we fake up an ACL_MASK entry.
Instead of adding fake ACL_MASK entries at the end, inserting them in the
correct position makes Solaris+VxFS happy. The Linux client and server
sides don't care about entry order. The three-entry-acl special case in
which we need a fake ACL_MASK entry was handled in xdr_nfsace_encode. The
patch moves this into nfsacl_encode.
Signed-off-by: Andreas Gruenbacher <agruen@suse.de>
Acked-by: Trond Myklebust <trond.myklebust@fys.uio.no>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
v9fs_file_read and v9fs_file_write use kmalloc to allocate buffers as big
as the data buffer received as parameter. kmalloc cannot be used to
allocate buffers bigger than 128K, so reading/writing data in chunks bigger
than 128k fails.
This patch reorganizes v9fs_file_read and v9fs_file_write to allocate only
buffers as big as the maximum data that can be sent in one 9P message.
Signed-off-by: Latchesar Ionkov <lucho@ionkov.net>
Cc: Eric Van Hensbergen <ericvh@gmail.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
In the current dell_rbu code ver 2.0 the packet update mechanism makes the
user app dump every individual packet in to the driver.
This adds in efficiency as every packet update makes the
/sys/class/firmware/dell_rbu/loading and data files to disappear and reappear
again. Thus the user app needs to wait for the files to reappear to dump
another packet. This slows down the packet update tremendously in case of
large number of packets. I am submitting a new patch for dell_rbu which will
change the way we do packet updates;
In the new method the user app will create a new single file which has already
packetized the rbu image and all the packets are now staged in this file.
This driver also creates a new entry in
/sys/devices/platform/dell_rbu/packet_size ; the user needs to echo the packet
size here before downloading the packet file.
The user should do the following:
create one single file which has all the packets stacked together.
echo the packet size in to /sys/devices/platform/dell_rbu/packet_size.
echo 1 > /sys/class/firmware/dell_rbu/loading
cat the packetfile > /sys/class/firmware/dell_rbu/data
echo 0 > /sys/class/firmware/dell_rbu/loading
The driver takes the file which came through /sys/class/firmware/dell_rbu/data
and takes chunks of paket_size data from it and place in contiguous memory.
This makes packet update process very efficient and fast. As all the packet
update happens in one single operation. The user can still read back the
downloaded file from /sys/devices/platform/dell_rbu/data.
Signed-off-by: Abhay Salunke <abhay_salunke@dell.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
pSeries_irq_bus_setup is marked __devinit but references s7a_workaround
which is marked __initdata.
Depending on who got the memory for s7a_workaround (and if the value was
now positive), it was possible for PCI hotplugged devices to have 3
subtracted from their interrupt number. This would happen randomly and
caused me much confusion :)
Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Search for a disconnect ccw_device on the ccw bus rather than on the css
bus (was a typo in patch I did for the klist conversion). A cast to an
embedding ccw_device from an embedded device in a struct subchannel will
lead us to oopses.
Signed-off-by: Cornelia Huck <cohuck@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Breakage noted by Al Viro.
It breaks non-PCI builds, it's probably better to have a more
direct implementation on sparc32, and which driver actually
needs this is still questionable.
We can resolve this in 2.6.15
Signed-off-by: David S. Miller <davem@davemloft.net>
Incorrect uart_write_wakeup() calls cause reference to a
NULL tty pointer in sunsab and sunzilog serial drivers.
Signed-off-by: David S. Miller <davem@davemloft.net>
Need to use long long, not long when RMWing a MSR. I think
it's harmless right now, but still should be better fixed
if AMD adds any bits in the upper 32bit of HWCR.
Bug was introduced with the TLB flush filter fix for i386
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
CPU hotplug fills up the possible map to NR_CPUs, but it did that after
setting up per CPU data. This lead to CPU data not getting allocated
for all possible CPUs, which lead to various side effects.
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
All the same issues - we can't just save the pointer to the thread, we
must save the pid/uid/euid combination.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
If a process issues an URB from userspace and (starts to) terminate
before the URB comes back, we run into the issue described above. This
is because the urb saves a pointer to "current" when it is posted to the
device, but there's no guarantee that this pointer is still valid
afterwards.
In fact, there are three separate issues:
1) the pointer to "current" can become invalid, since the task could be
completely gone when the URB completion comes back from the device.
2) Even if the saved task pointer is still pointing to a valid task_struct,
task_struct->sighand could have gone meanwhile.
3) Even if the process is perfectly fine, permissions may have changed,
and we can no longer send it a signal.
So what we do instead, is to save the PID and uid's of the process, and
introduce a new kill_proc_info_as_uid() function.
Signed-off-by: Harald Welte <laforge@gnumonks.org>
[ Fixed up types and added symbol exports ]
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
On the boot processor, we need to do the move onto the Linux trap
table a little bit differently else we'll take unhandlable faults in
the firmware address space.
Previously we would do the following:
1) Disable PSTATE_IE in %pstate.
2) Set %tba by hand to sparc64_ttable_tl0
3) Initialize alternate, mmu, and interrupt global
trap registers.
4) Call prom_set_traptable()
That doesn't work very well actually with the way we boot the kernel
VM these days. It worked by luck on many systems because the firmware
accesses for the prom_set_traptable() call happened to be loaded into
the TLB already, something we cannot assume.
So the new scheme is this:
1) Clear PSTATE_IE in %pstate and set %pil to 15
2) Call prom_set_traptable()
3) Initialize alternate, mmu, and interrupt global
trap registers.
and this works quite well. This sequence has been moved into a
callable function in assembler named setup-trap_table(). The idea is
that eventually trampoline.S can use this code as well. That isn't
possible currently due to some complications, but eventually we should
be able to do it.
Thanks to Meelis Roos for the Ultra5 boot failure report.
Signed-off-by: David S. Miller <davem@davemloft.net>
Noticed by Terence Ripperda
Undo wrong change in global_flush_tlb. We need to flush the caches in all
cases, not just when pages were reverted. This was a bogus optimization
added earlier, but it was wrong.
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Patch from Vincent Sanders
Add a defconfig for the ARM Collie platform
Signed-off-by: Richard Purdie <rpurdie@rpsys.net>
Signed-off-by: Vincent Sanders <vince@arm.linux.org.uk>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Patch from Vincent Sanders
Add a defconfig for the ARM Corgi Zarus platform
Signed-off-by: Richard Purdie <rpurdie@rpsys.net>
Signed-off-by: Vincent Sanders <vince@arm.linux.org.uk>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Patch from Vincent Sanders
Add a defconfig for the ARM Poodle Zarus platform
Signed-off-by: Richard Purdie <rpurdie@rpsys.net>
Signed-off-by: Vincent Sanders <vince@arm.linux.org.uk>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Patch from Vincent Sanders
Add a defconfig for the ARM Spitz Zarus platform
Signed-off-by: Richard Purdie <rpurdie@rpsys.net>
Signed-off-by: Vincent Sanders <vince@arm.linux.org.uk>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Patch from Nicolas Pitre
drivers/mfd/ucb1x00-core.c: In function 'ucb1x00_probe':
drivers/mfd/ucb1x00-core.c:482: error: 'ucb1x00_class' undeclared (first use in this function)
Signed-off-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
This fixes the setup of the alignment of the signal frame, so that all
signal handlers are run with a properly aligned stack frame.
The current code "over-aligns" the stack pointer so that the stack frame
is effectively always mis-aligned by 4 bytes. But what we really want
is that on function entry ((sp + 4) & 15) == 0, which matches what would
happen if the stack were aligned before a "call" instruction.
Signed-off-by: Markus F.X.J. Oberhumer <markus@oberhumer.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The logic in ide_do_request() doesn't guarantee that both drives will be
serviced after a call. It may "forget" to service one in some
circumstances, including when one of the drive is suspended (it will
eventually fail to service the slave when the master is suspended for
example). This prevents the wakeup requests that gets queued on wakeup
from sleep from beeing serviced in some cases when 2 drives are sharing
an IDE bus.
The problem is deep enough in the way this code works (and there are
probably a few other problematic but rare corner cases) and fixing it
would require some major rethinking of the way IDE decides which channel
to service. This is not 2.6.14 material. However, in the meantime,
Bart has accepted this simple workaround that will fix the crash on
wakeup from sleep since this specific corner case is actually hitting
users to get into 2.6.14.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The third param in this call to vmap shouldn't be GFP_KERNEL, which
makes no sense, but rather VM_MAP. Thanks to Al Viro for spotting
this.
Signed-off-by: Tom Zanussi <zanussi@us.ibm.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
UML/x86_64 doesn't run when built with frame pointers disabled. There
was an implicit frame pointer assumption in the stub segfault handler.
With frame pointers disabled, UML dies on handling its first page fault.
The container-of part of this is from Paolo Giarrusso <blaisorblade@yahoo.it>.
Signed-off-by: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The following patch makes swsusp avoid the possible temporary corruption
of page translation tables during resume on x86-64. This is achieved by
creating a copy of the relevant page tables that will not be modified by
swsusp and can be safely used by it on resume.
The problem is that during resume on x86-64 swsusp may temporarily
corrupt the page tables used for the direct mapping of RAM. If that
happens, a page fault occurs and cannot be handled properly, which leads
to the solid hang of the affected system. This leads to the loss of the
system's state from before suspend and may result in the loss of data or
the corruption of filesystems, so it is a serious issue. Also, it
appears to happen quite often (for me, as often as 50% of the time).
The problem is related to the fact that (at least) one of the PMD
entries used in the direct memory mapping (starting at PAGE_OFFSET)
points to a page table the physical address of which is much greater
than the physical address of the PMD entry itself. Moreover,
unfortunately, the physical address of the page table before suspend
(i.e. the one stored in the suspend image) happens to be different to
the physical address of the corresponding page table used during resume
(i.e. the one that is valid right before swsusp_arch_resume() in
arch/x86_64/kernel/suspend_asm.S is executed). Thus while the image is
restored, the "offending" PMD entry gets overwritten, so it does not
point to the right physical address any more (i.e. there's no page
table at the address pointed to by it, because it points to the address
the page table has been at during suspend). Consequently, if the PMD
entry is used later on, and it _is_ used in the process of copying the
image pages, a page fault occurs, but it cannot be handled in the normal
way and the system hangs.
In principle we can call create_resume_mapping() from
swsusp_arch_resume() (ie. from suspend_asm.S), but then the memory
allocations in create_resume_mapping(), resume_pud_mapping(), and
resume_pmd_mapping() must be made carefully so that we use _only_
NosaveFree pages in them (the other pages are overwritten by the loop in
swsusp_arch_resume()). Additionally, we are in atomic context at that
time, so we cannot use GFP_KERNEL. Moreover, if one of the allocations
fails, we should free all of the allocated pages, so we need to trace
them somehow.
All of this is done in the appended patch, except that the functions
populating the page tables are located in arch/x86_64/kernel/suspend.c
rather than in init.c. It may be done in a more elegan way in the
future, with the help of some swsusp patches that are in the works now.
[AK: move some externs into headers, renamed a function]
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Fix whitespace - I split this off the previous patch for easier review.
Signed-off-by: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
After restoring the existing code, make it work also when included in
kernelspace code (which isn't currently the case, but at least this will prevent
people from "fixing" it as just happened).
Whitespace is fixed in next patch - it cluttered the diff too much.
Signed-off-by: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Commit 44456d37b5, between 2.6.13-rc3 and -rc4,
was a "nice cleanup" which broke something. Revert the offending part.
It broke because:
a) because this part doesn't fall under the description
b) the author didn't know what he was doing here
c) the author didn't try to compile the existing code and see that it worked
perfectly.
d) the author didn't ask us what was happening
e) you didn't either, and somebody there should have learned that UML is a bit
different.
In fact, UML is special in linking to host libc and using its includes.
In particular, since host includes always define both __BIG_ENDIAN and
__LITTLE_ENDIAN, ntohll() macros started thinking to be in a big-endian world;
and on-disk compatibility was broken.
Many thanks go to Nix for reporting the problem and correctly diagnosing an
endianness problem.
Btw, this patch restores the previous code, which worked; but the definitions
would be uncorrect if used in kernelspace files.
Next patch addresses that.
Cc: Nix <nix@esperi.org.uk>, Olaf Hering <olh@suse.de>
Signed-off-by: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
For files which need to include glibc headers (i.e. userspace files), we
specified the correct flags only for .o, not for .s/.lst/.i. Fix this.
Signed-off-by: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Too many people were confused by skas0 and tried using "mode=skas0". And after
all, they are right - accept this.
Signed-off-by: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Add a missing $(Q) to a "ln" invocation.
Signed-off-by: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The pccardd thread has a race in it that it can shutdown in the
TASK_INTERRUPTIBLE state.
Make sure we mark ourselves runnable again as we remove ourselves from
the wait queue.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Patch from Ben Dooks
Initialise the driver's .owner field so that
the device driver can be referenced to the
module that owns it
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
