The TRACE_IRQS_ON function in iret_exc: calls a C function without
ensuring that the segments are set properly. Move the trace function and
the enabling of interrupt into the C stub.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Most of the time we can simply use the iret instruction to exit the
kernel, rather than having to use the iret hypercall - the only
exception is if we're returning into vm86 mode, or from delivering an
NMI (which we don't support yet).
When running native, iret has the behaviour of testing for a pending
interrupt atomically with re-enabling interrupts. Unfortunately
there's no way to do this with Xen, so there's a window in which we
could get a recursive exception after enabling events but before
actually returning to userspace.
This causes a problem: if the nested interrupt causes one of the
task's TIF_WORK_MASK flags to be set, they will not be checked again
before returning to userspace. This means that pending work may be
left pending indefinitely, until the process enters and leaves the
kernel again. The net effect is that a pending signal or reschedule
event could be delayed for an unbounded amount of time.
To deal with this, the xen event upcall handler checks to see if the
EIP is within the critical section of the iret code, after events
are (potentially) enabled up to the iret itself. If its within this
range, it calls the iret critical section fixup, which adjusts the
stack to deal with any unrestored registers, and then shifts the
stack frame up to replace the previous invocation.
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
This patch is a rollup of all the core pieces of the Xen
implementation, including:
- booting and setup
- pagetable setup
- privileged instructions
- segmentation
- interrupt flags
- upcalls
- multicall batching
BOOTING AND SETUP
The vmlinux image is decorated with ELF notes which tell the Xen
domain builder what the kernel's requirements are; the domain builder
then constructs the address space accordingly and starts the kernel.
Xen has its own entrypoint for the kernel (contained in an ELF note).
The ELF notes are set up by xen-head.S, which is included into head.S.
In principle it could be linked separately, but it seems to provoke
lots of binutils bugs.
Because the domain builder starts the kernel in a fairly sane state
(32-bit protected mode, paging enabled, flat segments set up), there's
not a lot of setup needed before starting the kernel proper. The main
steps are:
1. Install the Xen paravirt_ops, which is simply a matter of a
structure assignment.
2. Set init_mm to use the Xen-supplied pagetables (analogous to the
head.S generated pagetables in a native boot).
3. Reserve address space for Xen, since it takes a chunk at the top
of the address space for its own use.
4. Call start_kernel()
PAGETABLE SETUP
Once we hit the main kernel boot sequence, it will end up calling back
via paravirt_ops to set up various pieces of Xen specific state. One
of the critical things which requires a bit of extra care is the
construction of the initial init_mm pagetable. Because Xen places
tight constraints on pagetables (an active pagetable must always be
valid, and must always be mapped read-only to the guest domain), we
need to be careful when constructing the new pagetable to keep these
constraints in mind. It turns out that the easiest way to do this is
use the initial Xen-provided pagetable as a template, and then just
insert new mappings for memory where a mapping doesn't already exist.
This means that during pagetable setup, it uses a special version of
xen_set_pte which ignores any attempt to remap a read-only page as
read-write (since Xen will map its own initial pagetable as RO), but
lets other changes to the ptes happen, so that things like NX are set
properly.
PRIVILEGED INSTRUCTIONS AND SEGMENTATION
When the kernel runs under Xen, it runs in ring 1 rather than ring 0.
This means that it is more privileged than user-mode in ring 3, but it
still can't run privileged instructions directly. Non-performance
critical instructions are dealt with by taking a privilege exception
and trapping into the hypervisor and emulating the instruction, but
more performance-critical instructions have their own specific
paravirt_ops. In many cases we can avoid having to do any hypercalls
for these instructions, or the Xen implementation is quite different
from the normal native version.
The privileged instructions fall into the broad classes of:
Segmentation: setting up the GDT and the GDT entries, LDT,
TLS and so on. Xen doesn't allow the GDT to be directly
modified; all GDT updates are done via hypercalls where the new
entries can be validated. This is important because Xen uses
segment limits to prevent the guest kernel from damaging the
hypervisor itself.
Traps and exceptions: Xen uses a special format for trap entrypoints,
so when the kernel wants to set an IDT entry, it needs to be
converted to the form Xen expects. Xen sets int 0x80 up specially
so that the trap goes straight from userspace into the guest kernel
without going via the hypervisor. sysenter isn't supported.
Kernel stack: The esp0 entry is extracted from the tss and provided to
Xen.
TLB operations: the various TLB calls are mapped into corresponding
Xen hypercalls.
Control registers: all the control registers are privileged. The most
important is cr3, which points to the base of the current pagetable,
and we handle it specially.
Another instruction we treat specially is CPUID, even though its not
privileged. We want to control what CPU features are visible to the
rest of the kernel, and so CPUID ends up going into a paravirt_op.
Xen implements this mainly to disable the ACPI and APIC subsystems.
INTERRUPT FLAGS
Xen maintains its own separate flag for masking events, which is
contained within the per-cpu vcpu_info structure. Because the guest
kernel runs in ring 1 and not 0, the IF flag in EFLAGS is completely
ignored (and must be, because even if a guest domain disables
interrupts for itself, it can't disable them overall).
(A note on terminology: "events" and interrupts are effectively
synonymous. However, rather than using an "enable flag", Xen uses a
"mask flag", which blocks event delivery when it is non-zero.)
There are paravirt_ops for each of cli/sti/save_fl/restore_fl, which
are implemented to manage the Xen event mask state. The only thing
worth noting is that when events are unmasked, we need to explicitly
see if there's a pending event and call into the hypervisor to make
sure it gets delivered.
UPCALLS
Xen needs a couple of upcall (or callback) functions to be implemented
by each guest. One is the event upcalls, which is how events
(interrupts, effectively) are delivered to the guests. The other is
the failsafe callback, which is used to report errors in either
reloading a segment register, or caused by iret. These are
implemented in i386/kernel/entry.S so they can jump into the normal
iret_exc path when necessary.
MULTICALL BATCHING
Xen provides a multicall mechanism, which allows multiple hypercalls
to be issued at once in order to mitigate the cost of trapping into
the hypervisor. This is particularly useful for context switches,
since the 4-5 hypercalls they would normally need (reload cr3, update
TLS, maybe update LDT) can be reduced to one. This patch implements a
generic batching mechanism for hypercalls, which gets used in many
places in the Xen code.
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Chris Wright <chrisw@sous-sol.org>
Cc: Ian Pratt <ian.pratt@xensource.com>
Cc: Christian Limpach <Christian.Limpach@cl.cam.ac.uk>
Cc: Adrian Bunk <bunk@stusta.de>
The commit 635cf99a80 introduced a
regression. Executing a ptrace single step after certain int80
accesses will infinitely loop and never advance the PC.
The TIF_SINGLESTEP check should be done on the return from the syscall
and not before it.
I loops on each single step on the pop right after the int80 which writes out
to the console. At that point you can issue as many single steps as you want
and it will not advance any further.
The test case is below:
/* Test whether singlestep through an int80 syscall works.
*/
#define _GNU_SOURCE
#include <stdio.h>
#include <unistd.h>
#include <fcntl.h>
#include <sys/ptrace.h>
#include <sys/wait.h>
#include <sys/mman.h>
#include <asm/user.h>
#include <string.h>
static int child, status;
static struct user_regs_struct regs;
static void do_child()
{
char str[80] = "child: int80 test\n";
ptrace(PTRACE_TRACEME, 0, 0, 0);
kill(getpid(), SIGUSR1);
write(fileno(stdout),str,strlen(str));
asm ("int $0x80" : : "a" (20)); /* getpid */
}
static void do_parent()
{
unsigned long eip, expected = 0;
again:
waitpid(child, &status, 0);
if (WIFEXITED(status) || WIFSIGNALED(status))
return;
if (WIFSTOPPED(status)) {
ptrace(PTRACE_GETREGS, child, 0, ®s);
eip = regs.eip;
if (expected)
fprintf(stderr, "child stop @ %08lx, expected %08lx %s\n",
eip, expected,
eip == expected ? "" : " <== ERROR");
if (*(unsigned short *)eip == 0x80cd) {
fprintf(stderr, "int 0x80 at %08x\n", (unsigned int)eip);
expected = eip + 2;
} else
expected = 0;
ptrace(PTRACE_SINGLESTEP, child, NULL, NULL);
}
goto again;
}
int main(int argc, char * const argv[])
{
child = fork();
if (child)
do_parent();
else
do_child();
return 0;
}
Signed-off-by: Jason Wessel <jason.wessel@windriver.com>
Cc: Jeremy Fitzhardinge <jeremy@goop.org>
Cc: <stable@kernel.org>
Cc: Chuck Ebbert <76306.1226@compuserve.com>
Acked-by: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Convert VMI timer to use clock events, making it properly able to use the NO_HZ
infrastructure. On UP systems, with no local APIC, we just continue to route
these events through the PIT. On systems with a local APIC, or SMP, we provide
a single source interrupt chip which creates the local timer IRQ. It actually
gets delivered by the APIC hardware, but we don't want to use the same local
APIC clocksource processing, so we create our own handler here.
Signed-off-by: Zachary Amsden <zach@vmware.com>
Signed-off-by: Andi Kleen <ak@suse.de>
CC: Dan Hecht <dhecht@vmware.com>
CC: Ingo Molnar <mingo@elte.hu>
CC: Thomas Gleixner <tglx@linutronix.de>
Currently x86 (similar to x84-64) has a special per-cpu structure
called "i386_pda" which can be easily and efficiently referenced via
the %fs register. An ELF section is more flexible than a structure,
allowing any piece of code to use this area. Indeed, such a section
already exists: the per-cpu area.
So this patch:
(1) Removes the PDA and uses per-cpu variables for each current member.
(2) Replaces the __KERNEL_PDA segment with __KERNEL_PERCPU.
(3) Creates a per-cpu mirror of __per_cpu_offset called this_cpu_off, which
can be used to calculate addresses for this CPU's variables.
(4) Simplifies startup, because %fs doesn't need to be loaded with a
special segment at early boot; it can be deferred until the first
percpu area is allocated (or never for UP).
The result is less code and one less x86-specific concept.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Andi Kleen <ak@suse.de>
Xen wants a dedicated page for the GDT. I believe VMI likes it too.
lguest, KVM and native don't care.
Simple transformation to page-aligned "struct gdt_page".
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Andi Kleen <ak@suse.de>
Acked-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Fix a few clobbers to include the return register. The clobbers set
is the set of all registers modified (or may be modified) by the code
snippet, regardless of whether it was deliberate or accidental.
Also, make sure that callsites which are used in contexts which don't
allow clobbers actually save and restore all clobberable registers.
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Zachary Amsden <zach@vmware.com>
Some versions of libc can't deal with a VDSO which doesn't have its
ELF headers matching its mapped address. COMPAT_VDSO maps the VDSO at
a specific system-wide fixed address. Previously this was all done at
build time, on the grounds that the fixed VDSO address is always at
the top of the address space. However, a hypervisor may reserve some
of that address space, pushing the fixmap address down.
This patch does the adjustment dynamically at runtime, depending on
the runtime location of the VDSO fixmap.
[ Patch has been through several hands: Jan Beulich wrote the orignal
version; Zach reworked it, and Jeremy converted it to relocate phdrs
as well as sections. ]
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Zachary Amsden <zach@vmware.com>
Cc: "Jan Beulich" <JBeulich@novell.com>
Cc: Eric W. Biederman <ebiederm@xmission.com>
Cc: Andi Kleen <ak@suse.de>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Roland McGrath <roland@redhat.com>
Now we have an explicit per-cpu GDT variable, we don't need to keep the
descriptors around to use them to find the GDT: expose cpu_gdt directly.
We could go further and make load_gdt() pack the descriptor for us, or even
assume it means "load the current cpu's GDT" which is what it always does.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
No need to use -traditional for processing asm in i386/kernel/
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Annotate i386/kernel/entry.S with END/ENDPROC to assist disassemblers and
other analysis tools.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
VMI timer code. It works by taking over the local APIC clock when APIC is
configured, which requires a couple hooks into the APIC code. The backend
timer code could be commonized into the timer infrastructure, but there are
some pieces missing (stolen time, in particular), and the exact semantics of
when to do accounting for NO_IDLE need to be shared between different
hypervisors as well. So for now, VMI timer is a separate module.
[Adrian Bunk: cleanups]
Subject: VMI timer patches
Signed-off-by: Zachary Amsden <zach@vmware.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Andi Kleen <ak@suse.de>
Cc: Jeremy Fitzhardinge <jeremy@xensource.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Chris Wright <chrisw@sous-sol.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Convert the PDA code to use %fs rather than %gs as the segment for
per-processor data. This is because some processors show a small but
measurable performance gain for reloading a NULL segment selector (as %fs
generally is in user-space) versus a non-NULL one (as %gs generally is).
On modern processors the difference is very small, perhaps undetectable.
Some old AMD "K6 3D+" processors are noticably slower when %fs is used
rather than %gs; I have no idea why this might be, but I think they're
sufficiently rare that it doesn't matter much.
This patch also fixes the math emulator, which had not been adjusted to
match the changed struct pt_regs.
[frederik.deweerdt@gmail.com: fixit with gdb]
[mingo@elte.hu: Fix KVM too]
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Ian Campbell <Ian.Campbell@XenSource.com>
Acked-by: Ingo Molnar <mingo@elte.hu>
Acked-by: Zachary Amsden <zach@vmware.com>
Cc: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: Frederik Deweerdt <frederik.deweerdt@gmail.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
I wouldn't mind if CONFIG_COMPAT_VDSO went away entirely. But if it's there,
it should work properly. Currently it's quite haphazard: both real vma and
fixmap are mapped, both are put in the two different AT_* slots, sysenter
returns to the vma address rather than the fixmap address, and core dumps yet
are another story.
This patch makes CONFIG_COMPAT_VDSO disable the real vma and use the fixmap
area consistently. This makes it actually compatible with what the old vdso
implementation did.
Signed-off-by: Roland McGrath <roland@redhat.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
It has caused more problems than it ever really solved, and is
apparently not getting cleaned up and fixed. We can put it back when
it's stable and isn't likely to make warning or bug events worse.
In the meantime, enable frame pointers for more readable stack traces.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
It turns out that the most called ops, by several orders of magnitude,
are the interrupt manipulation ops. These are obvious candidates for
patching, so mark them up and create infrastructure for it.
The method used is that the ops structure has a patch function, which
is called for each place which needs to be patched: this returns a
number of instructions (the rest are NOP-padded).
Usually we can spare a register (%eax) for the binary patched code to
use, but in a couple of critical places in entry.S we can't: we make
the clobbers explicit at the call site, and manually clobber the
allowed registers in debug mode as an extra check.
And:
Don't abuse CONFIG_DEBUG_KERNEL, add CONFIG_DEBUG_PARAVIRT.
And:
AK: Fix warnings in x86-64 alternative.c build
And:
AK: Fix compilation with defconfig
And:
^From: Andrew Morton <akpm@osdl.org>
Some binutlises still like to emit references to __stop_parainstructions and
__start_parainstructions.
And:
AK: Fix warnings about unused variables when PARAVIRT is disabled.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Chris Wright <chrisw@sous-sol.org>
Signed-off-by: Zachary Amsden <zach@vmware.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Create a paravirt.h header for all the critical operations which need to be
replaced with hypervisor calls, and include that instead of defining native
operations, when CONFIG_PARAVIRT.
This patch does the dumbest possible replacement of paravirtualized
instructions: calls through a "paravirt_ops" structure. Currently these are
function implementations of native hardware: hypervisors will override the ops
structure with their own variants.
All the pv-ops functions are declared "fastcall" so that a specific
register-based ABI is used, to make inlining assember easier.
And:
+From: Andy Whitcroft <apw@shadowen.org>
The paravirt ops introduce a 'weak' attribute onto memory_setup().
Code ordering leads to the following warnings on x86:
arch/i386/kernel/setup.c:651: warning: weak declaration of
`memory_setup' after first use results in unspecified behavior
Move memory_setup() to avoid this.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Chris Wright <chrisw@sous-sol.org>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Jeremy Fitzhardinge <jeremy@goop.org>
Cc: Zachary Amsden <zach@vmware.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Andy Whitcroft <apw@shadowen.org>
The entry.S code at work_notifysig is surely wrong. It drops into unrelated
code if the branch to work_notifysig_v86 is taken, and CONFIG_VM86=n.
[PATCH] Make vm86 support optional
tree 9b5daef528
pushed to git Jan 8, 2006, and first appears in 2.6.16
The 'fix' here is to also compile out the vm86 test & branch when
CONFIG_VM86=n.
Signed-off-by: Joe Korty <joe.korty@ccur.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Use the cpu_number in the PDA to implement raw_smp_processor_id. This is a
little simpler than using thread_info, though the cpu field in thread_info
cannot be removed since it is used for things other than getting the current
CPU in common code.
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Chuck Ebbert <76306.1226@compuserve.com>
Cc: Zachary Amsden <zach@vmware.com>
Cc: Jan Beulich <jbeulich@novell.com>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
This patch is the meat of the PDA change. This patch makes several related
changes:
1: Most significantly, %gs is now used in the kernel. This means that on
entry, the old value of %gs is saved away, and it is reloaded with
__KERNEL_PDA.
2: entry.S constructs the stack in the shape of struct pt_regs, and this
is passed around the kernel so that the process's saved register
state can be accessed.
Unfortunately struct pt_regs doesn't currently have space for %gs
(or %fs). This patch extends pt_regs to add space for gs (no space
is allocated for %fs, since it won't be used, and it would just
complicate the code in entry.S to work around the space).
3: Because %gs is now saved on the stack like %ds, %es and the integer
registers, there are a number of places where it no longer needs to
be handled specially; namely context switch, and saving/restoring the
register state in a signal context.
4: And since kernel threads run in kernel space and call normal kernel
code, they need to be created with their %gs == __KERNEL_PDA.
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Chuck Ebbert <76306.1226@compuserve.com>
Cc: Zachary Amsden <zach@vmware.com>
Cc: Jan Beulich <jbeulich@novell.com>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Use asm-offsets for the offsets of registers into the pt_regs struct, rather
than having hard-coded constants
I left the constants in the comments of entry.S because they're useful for
reference; the code in entry.S is very dependent on the layout of pt_regs,
even when using asm-offsets.
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Keith Owens <kaos@ocs.com.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Clean up the espfix code:
- Introduced PER_CPU() macro to be used from asm
- Introduced GET_DESC_BASE() macro to be used from asm
- Rewrote the fixup code in asm, as calling a C code with the altered %ss
appeared to be unsafe
- No longer altering the stack from a .fixup section
- 16bit per-cpu stack is no longer used, instead the stack segment base
is patched the way so that the high word of the kernel and user %esp
are the same.
- Added the limit-patching for the espfix segment. (Chuck Ebbert)
[jeremy@goop.org: use the x86 scaling addressing mode rather than shifting]
Signed-off-by: Stas Sergeev <stsp@aknet.ru>
Signed-off-by: Andi Kleen <ak@suse.de>
Acked-by: Zachary Amsden <zach@vmware.com>
Acked-by: Chuck Ebbert <76306.1226@compuserve.com>
Acked-by: Jan Beulich <jbeulich@novell.com>
Cc: Andi Kleen <ak@muc.de>
Signed-off-by: Jeremy Fitzhardinge <jeremy@goop.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Current gcc generates calls not jumps to noreturn functions. When that happens the
return address can point to the next function, which confuses the unwinder.
This patch works around it by marking asynchronous exception
frames in contrast normal call frames in the unwind information. Then teach
the unwinder to decode this.
For normal call frames the unwinder now subtracts one from the address which avoids
this problem. The standard libgcc unwinder uses the same trick.
It doesn't include adjustment of the printed address (i.e. for the original
example, it'd still be kernel_math_error+0 that gets displayed, but the
unwinder wouldn't get confused anymore.
This only works with binutils 2.6.17+ and some versions of H.J.Lu's 2.6.16
unfortunately because earlier binutils don't support .cfi_signal_frame
[AK: added automatic detection of the new binutils and wrote description]
Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Andi Kleen <ak@suse.de>
We allow for the fact that the guest kernel may not run in ring 0. This
requires some abstraction in a few places when setting %cs or checking
privilege level (user vs kernel).
This is Chris' [RFC PATCH 15/33] move segment checks to subarch, except rather
than using #define USER_MODE_MASK which depends on a config option, we use
Zach's more flexible approach of assuming ring 3 == userspace. I also used
"get_kernel_rpl()" over "get_kernel_cs()" because I think it reads better in
the code...
1) Remove the hardcoded 3 and introduce #define SEGMENT_RPL_MASK 3 2) Add a
get_kernel_rpl() macro, and don't assume it's zero.
And:
Clean up of patch for letting kernel run other than ring 0:
a. Add some comments about the SEGMENT_IS_*_CODE() macros.
b. Add a USER_RPL macro. (Code was comparing a value to a mask
in some places and to the magic number 3 in other places.)
c. Add macros for table indicator field and use them.
d. Change the entry.S tests for LDT stack segment to use the macros
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Zachary Amsden <zach@vmware.com>
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Andi Kleen <ak@suse.de>
Abstract sensitive instructions in assembler code, replacing them with macros
(which currently are #defined to the native versions). We use long names:
assembler is case-insensitive, so if something goes wrong and macros do not
expand, it would assemble anyway.
Resulting object files are exactly the same as before.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Andi Kleen <ak@suse.de>
In i386's entry.S, FIX_STACK() needs annotation because it
replaces the stack pointer. And the rest of nmi() needs
annotation in order to compile with these new annotations.
Signed-off-by: Chuck Ebbert <76306.1226@compuserve.com>
Signed-off-by: Andi Kleen <ak@suse.de>
A kprobe executes IRET early and that could cause NMI recursion and stack
corruption.
Note: This problem was originally spotted and solved by Andi Kleen in the
x86_64 architecture. This patch is an adaption of his patch for i386.
AK: Merged with current code which was a bit different.
AK: Removed printk in nmi handler that shouldn't be there in the first time
AK: Added missing include.
AK: added KPROBES_END
Signed-off-by: Fernando Vazquez <fernando@intellilink.co.jp>
Signed-off-by: Andi Kleen <ak@suse.de>
And add proper CFI annotation to it which was previously
impossible. This prevents "stuck" messages by the dwarf2 unwinder
when reaching the top of a kernel stack.
Includes feedback from Jan Beulich
Cc: jbeulich@novell.com
Signed-off-by: Andi Kleen <ak@suse.de>
This patch moves the entry.S:error_entry to .kprobes.text section,
since code marked unsafe for kprobes jumps directly to entry.S::error_entry,
that must be marked unsafe as well.
This patch also moves all the ".previous.text" asm directives to ".previous"
for kprobes section.
AK: Following a similar i386 patch from Chuck Ebbert
AK: Also merged Jeremy's fix in.
+From: Jeremy Fitzhardinge <jeremy@goop.org>
KPROBE_ENTRY does a .section .kprobes.text, and expects its users to
do a .previous at the end of the function.
Unfortunately, if any code within the function switches sections, for
example .fixup, then the .previous ends up putting all subsequent code
into .fixup. Worse, any subsequent .fixup code gets intermingled with
the code its supposed to be fixing (which is also in .fixup). It's
surprising this didn't cause more havok.
The fix is to use .pushsection/.popsection, so this stuff nests
properly. A further cleanup would be to get rid of all
.section/.previous pairs, since they're inherently fragile.
+From: Chuck Ebbert <76306.1226@compuserve.com>
Because code marked unsafe for kprobes jumps directly to
entry.S::error_code, that must be marked unsafe as well.
The easiest way to do that is to move the page fault entry
point to just before error_code and let it inherit the same
section.
Also moved all the ".previous" asm directives for kprobes
sections to column 1 and removed ".text" from them.
Signed-off-by: Chuck Ebbert <76306.1226@compuserve.com>
Signed-off-by: Andi Kleen <ak@suse.de>
(And reset it on new thread creation)
It turns out that eflags is important to save and restore not just
because of iopl, but due to the magic bits like the NT bit, which we
don't want leaking between different threads.
Tested-by: Mike Galbraith <efault@gmx.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
CFA needs to be adjusted upwards for push, and downwards for pop.
arch/i386/kernel/entry.S gets it wrong in one place.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Acked-by: Jan Beulich <jbeulich@novell.com>
Cc: Andi Kleen <ak@muc.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Move the i386 VDSO down into a vma and thus randomize it.
Besides the security implications, this feature also helps debuggers, which
can COW a vma-backed VDSO just like a normal DSO and can thus do
single-stepping and other debugging features.
It's good for hypervisors (Xen, VMWare) too, which typically live in the same
high-mapped address space as the VDSO, hence whenever the VDSO is used, they
get lots of guest pagefaults and have to fix such guest accesses up - which
slows things down instead of speeding things up (the primary purpose of the
VDSO).
There's a new CONFIG_COMPAT_VDSO (default=y) option, which provides support
for older glibcs that still rely on a prelinked high-mapped VDSO. Newer
distributions (using glibc 2.3.3 or later) can turn this option off. Turning
it off is also recommended for security reasons: attackers cannot use the
predictable high-mapped VDSO page as syscall trampoline anymore.
There is a new vdso=[0|1] boot option as well, and a runtime
/proc/sys/vm/vdso_enabled sysctl switch, that allows the VDSO to be turned
on/off.
(This version of the VDSO-randomization patch also has working ELF
coredumping, the previous patch crashed in the coredumping code.)
This code is a combined work of the exec-shield VDSO randomization
code and Gerd Hoffmann's hypervisor-centric VDSO patch. Rusty Russell
started this patch and i completed it.
[akpm@osdl.org: cleanups]
[akpm@osdl.org: compile fix]
[akpm@osdl.org: compile fix 2]
[akpm@osdl.org: compile fix 3]
[akpm@osdl.org: revernt MAXMEM change]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Arjan van de Ven <arjan@infradead.org>
Cc: Gerd Hoffmann <kraxel@suse.de>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Zachary Amsden <zach@vmware.com>
Cc: Andi Kleen <ak@muc.de>
Cc: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Commit c3ff8ec31c ("[PATCH] i386: Don't
miss pending signals returning to user mode after signal processing")
meant that vm86 interrupt/signal handling got broken for the case when
vm86 is called from kernel space.
In this scenario, if signal is pending because of vm86 interrupt,
do_notify_resume/do_signal exits immediately due to user_mode() check,
without processing any signals. Thus, resume_userspace handler is spinning
in a tight loop with signal pending and TIF_SIGPENDING is set. Previously
everything worked Ok.
No in-tree usage of vm86() from kernel space exists, but I've heard
about a number of projects out there which use vm86 calls from kernel,
one of them being this, for instance:
http://dev.gentoo.org/~spock/projects/vesafb-tng/
The following patch fixes the issue.
Signed-off-by: Aleksey Gorelov <aleksey_gorelov@phoenix.com>
Cc: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Cc: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Remove the limit of 256 interrupt vectors by changing the value stored in
orig_{e,r}ax to be the complemented interrupt vector. The orig_{e,r}ax
needs to be < 0 to allow the signal code to distinguish between return from
interrupt and return from syscall. With this change applied, NR_IRQS can
be > 256.
Xen extends the IRQ numbering space to include room for dynamically
allocated virtual interrupts (in the range 256-511), which requires a more
permissive interface to do_IRQ.
Signed-off-by: Ian Pratt <ian.pratt@xensource.com>
Signed-off-by: Christian Limpach <Christian.Limpach@cl.cam.ac.uk>
Signed-off-by: Chris Wright <chrisw@sous-sol.org>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Cc: "Protasevich, Natalie" <Natalie.Protasevich@UNISYS.com>
Cc: Andi Kleen <ak@muc.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
To increase the usefulness of reliable stack unwinding, this adds CFI
unwind annotations to many low-level i386 routines.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
These are the i386-specific pieces to enable reliable stack traces. This is
going to be even more useful once CFI annotations get added to he assembly
code, namely to entry.S.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This adds an option to remove vm86 support under CONFIG_EMBEDDED. Saves
about 5k.
This version eliminates most of the #ifdefs of the previous version and
instead uses function stubs in vm86.h. Also, release_vm86_irqs is moved
from asm-i386/irq.h to a more appropriate home in vm86.h so that the stubs
can live together.
$ size vmlinux-baseline vmlinux-novm86
text data bss dec hex filename
2920821 523232 190652 3634705 377611 vmlinux-baseline
2916268 523100 190492 3629860 376324 vmlinux-novm86
Signed-off-by: Matt Mackall <mpm@selenic.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Mark some key kernel datastructures readonly. This patch was previously
posted on Jun 28th but was back then not merged because nothing was enforcing
rodata anyway.. well that changed now :)
Patch by Christoph Lameter <christoph@lameter.com> and Dave Jones
<davej@redhat.com>
Signed-off-by: Arjan van de Ven <arjan@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Cc: Andi Kleen <ak@muc.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patch contains the i386 architecture specific changes to prevent the
possible race conditions.
Signed-off-by: Prasanna S Panchamukhi <prasanna@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
As a follow-up to "UML Support - Ptrace: adds the host SYSEMU support, for
UML and general usage" (i.e. uml-support-* in current mm).
Avoid unconditionally jumping to work_pending and code copying, just reuse
the already existing resume_userspace path.
One interesting note, from Charles P. Wright, suggested that the API is
improvable with no downsides for UML (except that it will have to support
yet another host API, since dropping support for the current API, for UML,
is not reasonable from users' point of view).
Signed-off-by: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
CC: Charles P. Wright <cwright@cs.sunysb.edu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
With this patch, we change the way we handle switching from PTRACE_SYSEMU to
PTRACE_{SINGLESTEP,SYSCALL}, to free TIF_SYSCALL_EMU from double use as a
preparation for PTRACE_SYSEMU_SINGLESTEP extension, without changing the
behavior of the host kernel.
Signed-off-by: Bodo Stroesser <bstroesser@fujitsu-siemens.com>
Signed-off-by: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Cc: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Jeff Dike <jdike@addtoit.com>,
Paolo 'Blaisorblade' Giarrusso <blaisorblade_spam@yahoo.it>,
Bodo Stroesser <bstroesser@fujitsu-siemens.com>
Adds a new ptrace(2) mode, called PTRACE_SYSEMU, resembling PTRACE_SYSCALL
except that the kernel does not execute the requested syscall; this is useful
to improve performance for virtual environments, like UML, which want to run
the syscall on their own.
In fact, using PTRACE_SYSCALL means stopping child execution twice, on entry
and on exit, and each time you also have two context switches; with SYSEMU you
avoid the 2nd stop and so save two context switches per syscall.
Also, some architectures don't have support in the host for changing the
syscall number via ptrace(), which is currently needed to skip syscall
execution (UML turns any syscall into getpid() to avoid it being executed on
the host). Fixing that is hard, while SYSEMU is easier to implement.
* This version of the patch includes some suggestions of Jeff Dike to avoid
adding any instructions to the syscall fast path, plus some other little
changes, by myself, to make it work even when the syscall is executed with
SYSENTER (but I'm unsure about them). It has been widely tested for quite a
lot of time.
* Various fixed were included to handle the various switches between
various states, i.e. when for instance a syscall entry is traced with one of
PT_SYSCALL / _SYSEMU / _SINGLESTEP and another one is used on exit.
Basically, this is done by remembering which one of them was used even after
the call to ptrace_notify().
* We're combining TIF_SYSCALL_EMU with TIF_SYSCALL_TRACE or TIF_SINGLESTEP
to make do_syscall_trace() notice that the current syscall was started with
SYSEMU on entry, so that no notification ought to be done in the exit path;
this is a bit of a hack, so this problem is solved in another way in next
patches.
* Also, the effects of the patch:
"Ptrace - i386: fix Syscall Audit interaction with singlestep"
are cancelled; they are restored back in the last patch of this series.
Detailed descriptions of the patches doing this kind of processing follow (but
I've already summed everything up).
* Fix behaviour when changing interception kind #1.
In do_syscall_trace(), we check the status of the TIF_SYSCALL_EMU flag
only after doing the debugger notification; but the debugger might have
changed the status of this flag because he continued execution with
PTRACE_SYSCALL, so this is wrong. This patch fixes it by saving the flag
status before calling ptrace_notify().
* Fix behaviour when changing interception kind #2:
avoid intercepting syscall on return when using SYSCALL again.
A guest process switching from using PTRACE_SYSEMU to PTRACE_SYSCALL
crashes.
The problem is in arch/i386/kernel/entry.S. The current SYSEMU patch
inhibits the syscall-handler to be called, but does not prevent
do_syscall_trace() to be called after this for syscall completion
interception.
The appended patch fixes this. It reuses the flag TIF_SYSCALL_EMU to
remember "we come from PTRACE_SYSEMU and now are in PTRACE_SYSCALL", since
the flag is unused in the depicted situation.
* Fix behaviour when changing interception kind #3:
avoid intercepting syscall on return when using SINGLESTEP.
When testing 2.6.9 and the skas3.v6 patch, with my latest patch and had
problems with singlestepping on UML in SKAS with SYSEMU. It looped
receiving SIGTRAPs without moving forward. EIP of the traced process was
the same for all SIGTRAPs.
What's missing is to handle switching from PTRACE_SYSCALL_EMU to
PTRACE_SINGLESTEP in a way very similar to what is done for the change from
PTRACE_SYSCALL_EMU to PTRACE_SYSCALL_TRACE.
I.e., after calling ptrace(PTRACE_SYSEMU), on the return path, the debugger is
notified and then wake ups the process; the syscall is executed (or skipped,
when do_syscall_trace() returns 0, i.e. when using PTRACE_SYSEMU), and
do_syscall_trace() is called again. Since we are on the return path of a
SYSEMU'd syscall, if the wake up is performed through ptrace(PTRACE_SYSCALL),
we must still avoid notifying the parent of the syscall exit. Now, this
behaviour is extended even to resuming with PTRACE_SINGLESTEP.
Signed-off-by: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Cc: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Split the i386 entry.S files into entry.S and syscall_table.S which is
included in the previous one (so actually there is no difference between them)
and use the syscall_table.S in the UML build, instead of tracking by hand the
syscall table changes (which is inherently error-prone).
We must only insert the right #defines to inject the changes we need from the
i386 syscall table (for instance some different function names); also, we
don't implement some i386 syscalls, as ioperm(), nor some TLS-related ones
(yet to provide).
Signed-off-by: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
