cd21dfcfbb
Fix preemption and SMP problems in the FP emulator code.
...
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
cce812c99c
Ustat needs a wrapper on n32.
...
Signed-off-by: Thiemo Seufer <ths@networkno.de>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
1fcf1cc742
We pass a kernel pointer to do_sigaltstack in sys32_sigaltstack, so
...
we need to do the set_fs(KERNEL_DS) thing around this call.
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
3c37026d43
NPTL, round one.
...
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
f03da6e28e
Fix BogoMIPS display on UP and some minor cosmetical things.
...
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
589391a0fe
fcntl64 needs to be wrapped for n32.
...
Signed-off-by: Thiemo Seufer <ths@networkno.de>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
05b8042ac6
Fix one more case of computing the return EPC after the registers have
...
already been modified.
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
6dd0468852
When simulating ll/sc compute the return EPC before modifying the
...
registers.
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
2b07bd0235
Detect the 4KEcR2 and for now detect handle it like the 4KEc.
...
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
fe359bf584
Fixed buglet with previous patch that broke non au1x builds.
...
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
494900af68
Remove CONFIG_PM dependency from au1x wait in cpu_probe.
...
Additional work necessary to completely remove that config option.
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
1b3a6e975c
Fix 64bit SMP TLB handler and stack frame handling, optimize 32bit SMP
...
TLB handlers a bit, match definitions in pgtable-{32,64}.h better.
Signed-off-by: Thiemo Seufer <ths@networkno.de>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
209ac8ddb1
Use compat_sigval_t in struct compat_siginfo.
...
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
1592dac241
Reformatting, remove debugging code.
...
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
b4dbf95e30
Get rid of the the remains of 2.4-style ramdisk support.
...
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
71e0e556db
Multithreaded core dumps.
...
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
90a67b5909
sys_futex has 6 arguments.
...
Signed-off-by: Thiemo Seufer <ths@networkno.de>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
9ff77c469e
Export shm_align_mask and flush_data_cache_page.
...
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
77c728c224
Gcc 4.0 fixes.
...
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
cdaed73afb
Fix preemption bug.
...
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
fe00f943e0
Sparseify MIPS.
...
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
14f18b7f7e
On 24K we did always disable cache parity protection - obviously not
...
the greatest thing to do. Try to enable parity protection, check if
we actually succeeded and print a message about the outcome of this.
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
0f04afb595
ISOify.
...
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
e3ad1c23ba
Base Au1200 2.6 support.
...
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
784f7b9d89
Fix 'prctl' system call for IRIX. At this point IRIX 5.3 static binaries
...
are now working for 80% of the ones I have tried. The other ones that do
not work all fail in the same way with the same messages. Once that bug
is tracked down, we should be in good shape. Task locking still needs
some work.
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
8ab00b9a02
Convert struct hw_interrupt_type initializations to ISO C99 named
...
initializers.
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
b727a60258
Merge do_boot_cpu() into the new style __cpu_up().
...
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
0ac354801a
On multiprocessor systems the BogoMIPS for each CPU was reported was
...
the value for the last CPU having calibrated it's delay loop.
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
dc953df1ba
Fix wchan implementation, based on earlier by from Atsushi Nemoto.
...
Signed-off-by: Thiemo Seufer <ths@networkno.de>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
4e6a05fe5f
Improved modules loader, more robust and works on 64bit kernels.
...
Signed-off-by: Thiemo Seufer <ths@networkno.de>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
7ee8798f37
Until I figure out why NFS filesystems are having problems with
...
the 'load_irix_binary' and having kernel faults, Irix support is
disabled. I suspect locking of some sort, but I will now have to
investigate further.
Static IRIX binaries are now being detected properly and are using the
ELF interpreter found in this file.
Signed-off-by: Steven J. Hill <sjhill@realitydiluted.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
d1abb6a2b8
32-bit compatibility for various timer-related system calls.
...
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
a982099ca5
Update to match the native siginfo structure and code.
...
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
09276d905e
32-bit compatibility for ptrace GETEVENTMSG operation.
...
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
54f2da755b
Implement 32-bit compatibility for waitid(2).
...
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
a19050f301
Waitid(2) now has 5 arguments.
...
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
b6e203d84d
Use generic compat_sys_wait4 to implement 32-bit wait4(2).
...
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
57f0060b8a
Document why calling smp_call_function will deadlock when called with
...
interrupts disabled.
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
a18815abcd
Use preempt_schedule_irq.
...
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
925ddb04c5
Mask and ack CPU interrupts upon initialization. Keep the state
...
of software interrupts when unmasking.
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
c6237645d1
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
...
Fix TASK_SIZE for 32-bit processes on 64-bit kernels.
20 years ago
b053c98fbb
Fix register layout in o32 core dumps on 64-bit systems.
...
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
c83cfc9c94
Get rid of early_init. There's more need to make this form of
...
initialization actually useful and as is certainly unmergable with
upstream.
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
8c93650890
Dummy ISA DMA functions for systems that don't have ISA but share
...
drivers with ISA such as legacy free PCI.
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
b188ffe876
Fix build with SMP disabled and preemption enabled.
...
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
0964ce24d0
Move the invalid pmd and pte tables from .data to .bss. Fix alignment.
...
Signed-off-by: Thiemo Seufer <ths@networkno.de>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
c264852726
Remove unused arguments from preempt_{start,stop}/local_irq_{en,dis}able.
...
Don't clobber the preloaded TI_FLAGS in a2 needlessly.
Unexport local functions.
Signed-off-by: Thiemo Seufer <ths@networkno.de>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
69903d6500
Fix typos and formatting.
...
Signed-off-by: Thiemo Seufer <ths@networkno.de>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
b59a9504cb
De-optimize and decomplicate the spurious interrupt handler.
...
Signed-off-by: Thiemo Seufer <ths@networkno.de>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
20 years ago
2fd4ef85e0
[PATCH] error path in setup_arg_pages() misses vm_unacct_memory()
...
Pavel Emelianov and Kirill Korotaev observe that fs and arch users of
security_vm_enough_memory tend to forget to vm_unacct_memory when a
failure occurs further down (typically in setup_arg_pages variants).
These are all users of insert_vm_struct, and that reservation will only
be unaccounted on exit if the vma is marked VM_ACCOUNT: which in some
cases it is (hidden inside VM_STACK_FLAGS) and in some cases it isn't.
So x86_64 32-bit and ppc64 vDSO ELFs have been leaking memory into
Committed_AS each time they're run. But don't add VM_ACCOUNT to them,
it's inappropriate to reserve against the very unlikely case that gdb
be used to COW a vDSO page - we ought to do something about that in
do_wp_page, but there are yet other inconsistencies to be resolved.
The safe and economical way to fix this is to let insert_vm_struct do
the security_vm_enough_memory check when it finds VM_ACCOUNT is set.
And the MIPS irix_brk has been calling security_vm_enough_memory before
calling do_brk which repeats it, doubly accounting and so also leaking.
Remove that, and all the fs and arch calls to security_vm_enough_memory:
give it a less misleading name later on.
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-Off-By: Kirill Korotaev <dev@sw.ru>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
20 years ago
Ralf Baechle