Add the hypercall number to kvm_run and initialize it. This changes the ABI,
but as this particular ABI was unusable before this no users are affected.
Signed-off-by: Jeff Dike <[email protected]>
Signed-off-by: Avi Kivity <[email protected]>
This patch fixes a crash caused by an interrupt coming in when an IRQ stack
is being torn down. When this happens, handle_signal will loop, setting up
the IRQ stack again because the tearing down had finished, and handling
whatever signals had come in.
However, to_irq_stack returns a mask of pending signals to be handled, plus
bit zero is set if the IRQ stack was already active, and thus shouldn't be
torn down. This causes a problem because when handle_signal goes around
the loop, sig will be zero, and to_irq_stack will duly set bit zero in the
returned mask, faking handle_signal into believing that it shouldn't tear
down the IRQ stack and return thread_info pointers back to their original
values.
This will eventually cause a crash, as the IRQ stack thread_info will
continue pointing to the original task_struct and an interrupt will look
into it after it has been freed.
The fix is to stop passing a signal number into to_irq_stack. Rather, the
pending signals mask is initialized beforehand with the bit for sig already
set. References to sig in to_irq_stack can be replaced with references to
the mask.
[[email protected]: use UL]
Signed-off-by: Jeff Dike <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
The earlier crash dump fix on x86_64 depended on patches in -mm which
are intended for post-2.6.23. Without those, it broke the build when
it went into 2.6.23-rc5.
This changes the field references in ELF_CORE_COPY_REGS back to those
still used in mainline.
Signed-off-by: Jeff Dike <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Stop UML crashing when trying to dump a process core on x86_64. This is the
minimal fix to stop the crash - more things are broken here, and patches are
forthcoming.
The immediate thing to do is define ELF_CORE_COPY_REGS and
ELF_CORE_COPY_FPREGS. Defining ELF_CORE_COPY_FPREGS allows dump_fpu to go
away. It is defined in terms of save_fp_registers, so that needs to be added.
Signed-off-by: Jeff Dike <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Fix a class of bugs in the UML linker scripts which caused section boundary
variables to sometimes not line up with their sections.
Signed-off-by: Jeff Dike <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
The previous patch which limited the number of sectors in a single request
to a COWed device was correct in concept, but the limit was implemented in
the wrong place.
By putting it in ubd_add, it covered the cases where the COWing was
specified on the command line. However, when the command line only has the
COW file specified, the fact that it's a COW file isn't known until it's
opened, so the limit is missed in these cases.
This patch moves the sector limit from ubd_add to ubd_open_dev.
Signed-off-by: Jeff Dike <[email protected]>
Cc: <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
The previous DEBUG_SHIRQ patch missed one case. The console doesn't
set its host descriptors non-blocking.
Signed-off-by: Jeff Dike <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
2.6.23-rc1 turned up another batch of references from non-__init code to
__init code. In most cases, these were missing __init annotations. In one
case (os_drop_memory), the annotation was present but wrong.
init_maps is __init, but for some reason was being very careful about the
mechanism by which it allocated memory, checking whether it was OK to use
kmalloc (at this point in the boot, it definitely isn't) and using either
alloc_bootmem_low_pages or kmalloc/vmalloc. So, the kmalloc/vmalloc code is
removed.
Signed-off-by: Jeff Dike <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Restructure do_aio thanks to commments from Ulrich and Al.
Uli started this by seeing that UML's initialization of a struct iocb
initialized fields that it shouldn't.
Al followed up by adding the following cleanups:
eliminating a variable by just using an anonymous structure in
its place.
hoisting a duplicated line out of the switch.
simplifying the error checking at the end.
I added a severity to the printk.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Al Viro <[email protected]>
Cc: Ulrich Drepper <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
In 2.6.23-rc1, i386 fiddled its string support such that UML started getting
undefined references from modules. The UML asm/string.h was including the
i386 string.h, which defined __HAVE_ARCH_STR*, but the corresponding
implementations weren't being pulled in.
This is fixed by adding arch/i386/lib/string.h to the list of host
architecture files to be pulled in to UML.
A complication is that the libc exports file assumed that the generic strlen
and strstr weren't in use (i.e. __HAVE_ARCH_STR is defined), then they aren't
exported. This is untrue for strlen, which is exported in either case, so
this logic is not needed.
Signed-off-by: Jeff Dike <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
I forgot this file was here. It hasn't been used since UML has been in
mainline.
Thanks to Jesper for finding something that needed doing to it, thus reminding
me of its existence.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
COWed devices can't handle more than 32 (64 on x86_64) sectors in one request
due to the size of the bitmap being carried around in the io_thread_req.
Enforce that by telling the block layer not to put too many sectors in
requests to COWed devices.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Cc: <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Add some exports for hostfs that are required after Alberto Bertogli's fixes
for accessing unlinked host files.
Also did some style cleanups while I was here.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
UML had two wrapper procedures for kmalloc, um_kmalloc and um_kmalloc_atomic
because the flag constants weren't available in userspace code.
kern_constants.h had made kernel constants available for a long time, so there
is no need for these wrappers any more. Rather, userspace code calls kmalloc
directly with the userspace versions of the gfp flags.
kmalloc isn't a real procedure, so I had to essentially copy the inline
wrapper around __kmalloc.
vmalloc also had its own wrapper for no good reason. This is now gone.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
run_helper and run_helper_thread had arguments which were the same in all
callers. run_helper's stack_out was always NULL and run_helper_thread's
stack_order was always 0. These are now gone, and the constants folded
into the code.
Also fixed leaks of the helper stack in the AIO and SIGIO code.
Signed-off-by: Jeff Dike <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Cleanup of the SIGWINCH support.
Some code and comment reformatting.
The stack used for SIGWINCH threads was leaked. This is now fixed by storing
it with the pid and other information, and freeing it when the thread is
killed.
If something goes wrong with a WIGWINCH thread, and this is discovered in the
interrupt handler, the winch record would leak. It is now freed, except that
the IRQ isn't freed. This is hard to do from interrupt context. This has the
side-effect that the IRQ system maintains a reference to the freed structure,
but that shouldn't cause a problem since the descriptor is disabled.
register_winch_irq is now much better about cleaning up after an
initialization failure.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
If the host side of a console can't be opened, this will now produce visible
error messages.
enable_chan now returns a status and this is passed up to con_open and
ssl_open, which will complain if anything went wrong.
The default host device for the serial line driver is now a pts device rather
than a pty device since lots of hosts have LEGACY_PTYS disabled. This had
always been failing on such hosts, but silently.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Cleanup, mostly style violations.
Tidied the includes.
getmaster returns a real errno, which pty_open returns if there's a
problem.
The printks now have severity.
Changed os_* calls to call libc directly.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Major tidying of the xterm console driver:
got rid of the tt-mode gdb support
tidied up the includes
fixed lots of style violations
replaced os_* calls with glibc calls in xterm.c
all printk calls now have a severity indicator
the error paths of xterm_open are closer to being right
Signed-off-by: Jeff Dike <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
For some reason, I was using kmalloc instead of get_free_pages for kernel
stacks.
Signed-off-by: Jeff Dike <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
It is theoretically possible for a request to finish and be freed between
writing it to the I/O thread and updating the sector count. In this case, the
update will dereference a freed pointer.
To avoid this, I delay the update until processing the next sg segment, when
the request pointer is known to be good.
Signed-off-by: Jeff Dike <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Add asm-um/paravirt.h so that i386 headers that get pulled into UML
don't cause build failures when they want asm/paravirt.h.
Signed-off-by: Jeff Dike <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
The x86_64 a.out.h got a definition of STACK_TOP_MAX, which interferes with
the UML version. So, just undef it like STACK_TOP.
Signed-off-by: Jeff Dike <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Distros seem to be removing PAGE_SIZE from asm/page.h. So, the libc side of
UML should stop using it.
I replace it with UM_KERN_PAGE_SIZE, which is defined to be the same as
PAGE_SIZE on the kernel side of the house. I could also use getpagesize(),
but it's more important that UML have the same value of PAGE_SIZE everywhere.
It's conceivable that it could be built with a larger PAGE_SIZE, and use of
getpagesize() would break that badly.
PAGE_MASK got the same treatment, as it is closely tied to PAGE_SIZE.
Signed-off-by: Jeff Dike <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Include linux/kernel.h wherever simple_strtoul is used. This kills a
compile warning in stderr_console.c and potential ones in the other files.
This also fixes a bunch of style violations in exitcode.c.
Signed-off-by: Jeff Dike <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Force KERNEL_STACK_ORDER to be at least 1 on UML/x86_64, to avoid overflows.
Signed-off-by: Jeff Dike <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Make the PTRACE_SYSEMU checking more robust. It will make sure that system
call numbers are reported correctly. If there is a problem, it will disable
PTRACE_SYSEMU use and use PTRACE_SYSCALL instead.
Thanks to Balaji G for helping reproduce this problem.
Signed-off-by: Jeff Dike <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Make kernel stacks be 1 page on i386 and 2 pages on x86_64. These match the
host values.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Add a separate IRQ stack. This differs from i386 in having the entire
interrupt run on a separate stack rather than starting on the normal kernel
stack and switching over once some preparation has been done. The underlying
mechanism, is of course, sigaltstack.
Another difference is that interrupts that happen in userspace are handled on
the normal kernel stack. These cause a wait wakeup instead of a signal
delivery so there is no point in trying to switch stacks for these. There's
no other stuff on the stack, so there is no extra stack consumption.
This quirk makes it possible to have the entire interrupt run on a separate
stack - process preemption (and calls to schedule()) happens on a normal
kernel stack. If we enable CONFIG_PREEMPT, this will need to be rethought.
The IRQ stack for CPU 0 is declared in the same way as the initial kernel
stack. IRQ stacks for other CPUs will be allocated dynamically.
An extra field was added to the thread_info structure. When the active
thread_info is copied to the IRQ stack, the real_thread field points back to
the original stack. This makes it easy to tell where to copy the thread_info
struct back to when the interrupt is finished. It also serves as a marker of
a nested interrupt. It is NULL for the first interrupt on the stack, and
non-NULL for any nested interrupts.
Care is taken to behave correctly if a second interrupt comes in when the
thread_info structure is being set up or taken down. I could just disable
interrupts here, but I don't feel like giving up any of the performance gained
by not flipping signals on and off.
If an interrupt comes in during these critical periods, the handler can't run
because it has no idea what shape the stack is in. So, it sets a bit for its
signal in a global mask and returns. The outer handler will deal with this
signal itself.
Atomicity is had with xchg. A nested interrupt that needs to bail out will
xchg its signal mask into pending_mask and repeat in case yet another
interrupt hit at the same time, until the mask stabilizes.
The outermost interrupt will set up the thread_info and xchg a zero into
pending_mask when it is done. At this point, nested interrupts will look at
->real_thread and see that no setup needs to be done. They can just continue
normally.
Similar care needs to be taken when exiting the outer handler. If another
interrupt comes in while it is copying the thread_info, it will drop a bit
into pending_mask. The outer handler will check this and if it is non-zero,
will loop, set up the stack again, and handle the interrupt.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Some tidying of the irq code before introducing irq stacks. Mostly
style fixes, but the timer handler calls the timer code directly
rather than going through the generic sig_handler_common_skas.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Now that we have UM_THREAD_SIZE, we can replace the calculations in
user-space code (an earlier patch took care of the kernel side of the
house).
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Replaced task_protections with stack_protections since they do the same
thing, and task_protections was misnamed anyway.
This needs THREAD_SIZE, so that's imported via common-offsets.h
Also tidied up the code in the vicinity.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
A linuxdoc comment had fallen out of date - it refers to an argument which no
longer exists.
Signed-off-by: Jeff Dike <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Allow a pcap device to be assigned a MAC on the command line. They don't
really need one, but it is handy to be able to do when your distro assigns a
new ethernet device whenever it sees a new MAC.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Some network device cleanup.
When setup_etheraddr found a globally valid MAC being assigned to an
interface, it went ahead and used it rather than assigning a random MAC like
the other cases do. This isn't really an error like the others, but it seems
consistent to make it behave the same.
We were getting some duplicate kfree() in the error case in eth_configure
because platform_device_unregister frees buffers that the error cases
following tried to free again.
The pcap initialization routine wasn't doing the proper printk of its
information, causing a printk of the first part of that line to be
unterminated by a newline.
The pcap code had a bunch of style violations, which are now fixed.
pcap_setup wasn't returning false when it detected an unrecognized
option.
The printks in pcap_user all got UM_KERN_BLAH prepended to their
format strings.
pcap_remove now checks for a non-NULL pcap structure before it calls
pcap_close.
Signed-off-by: Jeff Dike <[email protected]>
Acked-by: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
i386:
Rearrange the cmpxchg code to allow atomic.h to get it without needing to
include system.h. This kills warnings in the UML build from atomic.h about
implicit declarations of cmpxchg symbols. The i386 build presumably isn't
seeing this because a separate inclusion of system.h is covering it over.
The cmpxchg stuff is moved to asm-i386/cmpxchg.h, with an include left in
system.h for the benefit of generic code which expects cmpxchg there.
Meanwhile, atomic.h includes cmpxchg.h.
This causes no noticable damage to the i386 build.
x86_64:
Move cmpxchg into its own header. atomic.h already included system.h, so
this is changed to include cmpxchg.h.
This is purely cleanup - it's not fixing any warnings - so if the x86_64
system.h isn't considered as cleanup-worthy as i386, then this can be
dropped.
It causes no noticable damage to the x86_64 build.
uml:
The i386 and x86_64 cmpxchg patches require an asm-um/cmpxchg.h for the
UML build.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Cc: Andi Kleen <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
The ever-vigilant users of linode.com noticed that an idle 2.6 UML has a
persistent load average of ~.4.
It turns out that because the UML timer handler processed softirqs before
actually delivering the tick, the tick was counted in the context of the idle
thread about half the time.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
With the current timekeeping, !CONFIG_UML_REAL_TIME_CLOCK has
inconsistent behavior. Previously, gettimeofday could be (and was)
isolated from the clock ticking. Now, it's not, so when
CONFIG_UML_REAL_TIME_CLOCK is disabled, gettimeofday must progress in
lockstep with the clock, making it fully virtual.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
It turns out that the message complaining about a lack of tmpfs space
on the host can be misunderstood as referring to the UML.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
When doing a full address space flush, only look at areas covered by a VMA.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
More trimming of the page fault path.
Permissions are passed around in a single int rather than one bit per
int. The permission values are copied from libc so that they can be
passed to mmap and mprotect without any further conversion.
The register sets used by do_syscall_stub and copy_context_skas0 are
initialized once, at boot time, rather than once per call.
wait_stub_done checks whether it is getting the signals it expects by
comparing the wait status to a mask containing bits for the signals of
interest rather than comparing individually to the signal numbers. It
also has one check for a wait failure instead of two. The caller is
expected to do the initial continue of the stub. This gets rid of an
argument and some logic. The fname argument is gone, as that can be
had from a stack trace.
user_signal() is collapsed into userspace() as it is basically one or
two lines of code afterwards.
The physical memory remapping stuff is gone, as it is unused.
flush_tlb_page is inlined.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Give the page fault code a specialized path. There is only one page to look
at, so there's no point in going into the general page table walking code.
There's only going to be one host operation, so there are no opportunities for
merging. So, we go straight to the pte we want, figure out what needs doing,
and do it.
While I was in here, I fixed the wart where the address passed to unmap was a
void *, but an unsigned long to map and protect.
This gives me just under 10% on a kernel build.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Allow deadlocks to be avoided in the AIO code by setting the pipe to the I/O
thread non-blocking.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Rename os_{read_write}_file_k back to os_{read_write}_file, delete
the originals and their bogus infrastructure, and fix all the callers.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Formatting fixes ahead of renaming os_{read_write}_file_k to
os_{read_write}_file and fixing all the callers.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Convert all remaining os_{read_write}_file users to use the simple
{read,write} wrappers, os_{read_write}_file_k.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Code running on the initial UML stack can't receive or process signals since
current must be valid when IRQs are handled, and there is no current for this
stack.
So, instead of using UML_LONGJMP and UML_SETJMP, which are careful to save and
restore signal state, and, as a side-effect, handle any deferred signals,
start_idle_thread must use the bare equivalents, which don't do anything with
signals.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Dump core after a panic. This will provide better debugging information than
is currently available.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Instead of writing entire structures between UML and the I/O thread, we send
pointers. This cuts down on the amount of data being copied and possibly
allows more requests to be pending between the two.
This requires that the requests be kmalloced and freed instead of living on
the stack.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Send as many I/O requests to the I/O thread as possible, even though it will
still only handle one at a time. This provides an opportunity to reduce
latency by starting one request before the previous one has been finished in
the driver.
Request handling is somewhat modernized by requesting sg pieces of a request
and handling them separately, finishing off the entire request after all the
pieces are done.
When a request queue stalls, normally because its pipe to the I/O thread is
full, it is put on the restart list. This list is processed by starting up
the queues on it whenever there is some indication that progress might be
possible again. Currently, this happens in the driver interrupt routine.
Some requests have been finished, so there is likely to be room in the pipe
again.
This almost doubles throughput when copying data between devices, but made no
noticable difference on anything else I tried.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Cc: Jens Axboe <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
This patch converts calls in the os layer to os_{read,write}_file to calls
directly to libc read() and write() where it is clear that the I/O buffer is
in the kernel.
We can do that here instead of calling os_{read,write}_file_k since we are in
libc code and can call libc directly.
With the change in the calls, error handling needs to be changed to refer to
errno directly rather than the return value of the call.
CATCH_EINTR wrappers were also added where needed.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
This patch lays some groundwork for the next one, which converts calls to
os_{read,write}_file into {read,write}, by doing some tidying in the affected
areas.
do_not_aio gets restructured to make the final result a bit cleaner.
There are also whitespace and other formatting fixes, fixes in error messages,
and a typo fix.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
This patch starts the removal of a very old, very broken piece of code. This
stems from the problem of passing a userspace buffer into read() or write() on
the host. If that buffer had not yet been faulted in, read and write will
return -EFAULT.
To avoid this problem, the solution was to fault the buffer in before the
system call by touching the pages that hold the buffer by doing a copy-user of
a byte to each page. This is obviously bogus, but it does usually work, in tt
mode, since the kernel and process are in the same address space and userspace
addresses can be accessed directly in the kernel.
In skas mode, where the kernel and process are in separate address spaces, it
is completely bogus because the userspace address, which is invalid in the
kernel, is passed into the system call instead of the corresponding physical
address, which would be valid. Here, it appears that this code, on every host
read() or write(), tries to fault in a random process page. This doesn't seem
to cause any correctness problems, but there is a performance impact. This
patch, and the ones following, result in a 10-15% performance gain on a kernel
build.
This code can't be immediately tossed out because when it is, you can't log
in. Apparently, there is some code in the console driver which depends on
this somehow.
However, we can start removing it by switching the code which does I/O using
kernel addresses to using plain read() and write(). This patch introduces
os_read_file_k and os_write_file_k for use with kernel buffers and converts
all call locations which use obvious kernel buffers to use them. These
include I/O using buffers which are local variables which are on the stack or
kmalloc-ed. Later patches will handle the less obvious cases, followed by a
mass conversion back to the original interface.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
It turns out that essentially none of the x86_64 bugs.c is needed. So, we can
delete most of it.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
The previous page table walking code was horribly inefficient. This patch
replaces it with code taken from elsewhere in the kernel.
Forking from bash is now ~5% faster and page faults are handled ~10% faster.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Provide a register dump if handle_trap fails. Abstract out ptrace_dump_regs
since it now has two callers.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Define release methods for the ubd and net drivers. They contain as much of
the remove methods as make sense. All error checking must have already been
done as well as anything else that might be holding a reference on the device
kobject.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
HOST_FRAME_SIZE isn't used any more. It has been replaced with MAX_REG_NR.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Commentary about missing locking.
Also got rid of uml_start because it was pointless.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
If there's a segfault inside the kernel, we want a dump of the registers at
the point of the segfault, not the registers at the point of calling panic or
the last userspace registers.
sig_handler_common_skas now uses a static register set in the case of a
SIGSEGV to avoid messing up the process registers if the segfault turns out to
be non-fatal.
The architecture sigcontext-to-pt_regs copying code was repurposed to copy
data out of the SEGV stack frame.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Tidying in preparation for the segfault register dumping patch which follows.
void * pointers are changed to union uml_pt_regs *. This makes the types
match reality, except in arch_fixup, which is changed to operate on a union
uml_pt_regs. This fixes a bug in the call from segv_handler, which passes a
union uml_pt_regs, to segv, which expects to pass a struct sigcontext to
arch_fixup.
Whitespace and other style fixes.
There's also a errno printk fix.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
kernel_thread() should just return an error value on do_fork failure, not
panic.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
userspace code used to have to call the kernelspace function page_size() in
order to determine the value of the kernel's PAGE_SIZE. Since this is now
available directly from kern_constants.h as UM_KERN_PAGE_SIZE, page_size() can
be deleted and calls changed to use the constant.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Clean up arch/um/kernel/process.c:
- lots of return(x); -> return x; conversions
- a number of the small functions are either unused, in which case they are
gone, along any declarations in a header, or could be made static.
- current_pid is ifdefed on CONFIG_MODE_TT and its declaration is ifdefed on
both CONFIG_MODE_TT and UML_CONFIG_MODE_TT because we don't know whether
it's being used in a userspace or kernel file.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Comment the lack of locking on a couple of globals.
Also fix the formatting of __setup_host_supports_tls.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
flush_thread doesn't need to do a full page table walk in order to clear the
address space. It knows what the end result needs to be, so it can call unmap
directly.
This results in a 10-20% speedup in an exec from bash.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
The build started finding calls from non-init to init functions. These are
just cases of init functions not being properly marked, so this patch fixes
that.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
user_util.h isn't needed any more, so delete it and remove all includes of it.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Rescue the useful contents of the soon-to-be-gone user-util.h.
pty.c now gets ptsname from stdlib.h like it should have always done.
CATCH_EINTR is now in os.h, although perhaps all usage should be under
os-Linux at some point.
get_pty is also in os.h.
This patch restores the old definition of ARRAY_SIZE in user.h. This file is
included only in userspace files, so there will be no conflict with the
kernel's new ARRAY_SIZE. The copy of the kernel's ARRAY_SIZE and associated
infrastructure is now gone.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
This patch moves all the the symbols defined in um_arch.c, which are mostly
boundaries between different parts of the UML kernel address space, to a new
header, as-layout.h. There are also a few things here which aren't really
related to address space layout, but which don't really have a better place to
go.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
This patch moves the declarations of the architecture hooks from user_util.h
to a new header, arch.c, and adds the necessary includes to files which need
those declarations.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
This patch narrows the sigio interface. The boot-time SIGIO testing used to
be in start_up.c, which meant that pty_output_sigio and pty_close_sigio needed
to be global. By moving that code here, those can become static and the
declarations moved from user_util.h.
os_check_bugs is also here because it only does the SIGIO checking. If it
does more, it'll probably move back to start_up.c.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
This fixes a number of problems associated with network interface hotplug.
The userspace initialization function can fail in some cases, but the
failure was never passed back to eth_configure, which proceeded with the
configuration. This results in a zombie device that is present, but can't
work. This is fixed by allowing the initialization routines to return an
error, which is checked, and the configuration aborted on failure.
eth_configure failed to check for many failures. Even when it did check,
it didn't undo whatever initializations has already happened, so a present,
but partially initialized and non-working device could result. It now
checks everything that can fail, and bails out, undoing whatever had been
done.
The return value of eth_configure was always ignored, so it is now just
void.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Cc: Jeff Garzik <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Fix a bunch of formatting violations in the drivers:
return(n) -> return n
whitespace fixes
emacs formatting comment removal
breaking if(foo) return(n) into two lines
There are also a couple of errno use bugs:
using errno in a printk when the failure put errno into a local variable
saving errno after a printk, which can change it
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
If a disk fails to open, i.e. its host file doesn't exist, it won't be
removable because the hot-unplug code checks the existence of its gendisk.
This won't exist because it is only allocated for successfully opened disks.
Thus, a typo on the command line can result in a unusable and unfixable disk.
This is fixed by freeing the gendisk if it's there, but not letting that
affect the removal.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Print out core dump limits at boot time. This is to allow core dumps
to be collected if something goes very wrong and to tell if a core
dump isn't going to happen because of a resource limit.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Move the host_info string from util.c to um_arch.c, where it is
actually initialized and used. Also document its lack of locking.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Get rid of a bunch of unused stuff -
cpu_feature had no users
linux_prog is little-used, so its declaration is moved to the
user for easy deletion when the whole file goes away
a long-unused debugging aid in helper.c is gone
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Fix a few miscellaneous compilation problems -
an assignment with mismatched types in ldt.c
a missing include in mconsole.h which needs a definition of uml_pt_regs
I missed removing an include of user_util.h in hostfs
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
As the comment immediately preceding this points out, this list is changed in
irq context, so it needs to be protected with spin_lock_irqsave in process
context when it is processed.
Sometimes, gcc should just compile the comments and forget the code.
The IRQ side of this was better, in the sense that it blocked and unblocked
interrupts, but it still should have saved and restored them.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Fix a NULL dereference when unplugging a device. The default value of
err_msg wants to be "" in case the driver doesn't modify it.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Commit 62f96cb01e introduced per-devices
queues and locks, which was fine as far as it went, but left in place a
global which controlled access to submitting requests to the host. This
should have been made per-device as well, since it causes I/O hangs when
multiple block devices are in use.
This patch fixes that by replacing the global with an activity flag in the
device structure in order to tell whether the queue is currently being run.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
This patch uses MAX_REG_NR consistently to refer to the register file size.
FRAME_SIZE isn't sufficient because on x86_64, it is smaller than the
ptrace register file size. MAX_REG_NR was introduced as a consistent way
to get the number of registers, but wasn't used everywhere it should be.
When this causes a problem, it makes PTRACE_SETREGS fail on x86_64 because
of a corrupted segment register value in the known-good register file. The
patch also adds a register dump at that point in case there are any future
problems here.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Cc: <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
During a static link, ld has started putting a .note section in the
.uml.setup.init section. This has the result that the UML setups begin
with 32 bytes of garbage and UML crashes immediately on boot.
This patch creates a specific .note section for ld to drop this stuff
into.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Cc: <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
In my previous x86_64 thread fix, I forgot to initialize thread.arch.fs in
arch_prctl. A process calling arch_prctl to set %fs would lose it on the
next context switch.
It also turns out that you can switch to a process which is in the process
of exiting and which has lost its mm. In this case, it's worse than
useless to try to call arch_prctl on the host process.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Comment the fact that sig_info is initialized early in boot, and thus doesn't
need any locking.
Signed-off-by: Jeff Dike <[email protected]>
Cc: Paolo 'Blaisorblade' Giarrusso <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Jeff Dike