linux.git/arch/mips/kernel, branch v2.6.21.7

[MIPS] Fix BUG(), BUG_ON() handling

2007-04-20T13:58:37+00:00

With commit 63dc68a8cf60cb110b147dab1704d990808b39e2, kernel can not
handle BUG() and BUG_ON() properly since get_user() returns false for
kernel code.  Use __get_user() to skip unnecessary access_ok().  This
patch also make BRK_BUG code encoded in the TNE instruction.

Signed-off-by: Atsushi Nemoto 
Signed-off-by: Ralf Baechle

[MIPS] Retry {save,restore}_fp_context if failed in atomic context.

2007-04-20T13:58:37+00:00

The save_fp_context()/restore_fp_context() might sleep on accessing
user stack and therefore might lose FPU ownership in middle of them.

If these function failed due to "in_atomic" test in do_page_fault,
touch the sigcontext area in non-atomic context and retry these
save/restore operation.

This is a replacement of a (broken) fix which was titled "Allow CpU
exception in kernel partially".

Signed-off-by: Atsushi Nemoto 
Signed-off-by: Ralf Baechle

[MIPS] Disallow CpU exception in kernel again.

2007-04-20T13:58:37+00:00

The commit 4d40bff7110e9e1a97ff8c01bdd6350e9867cc10 ("Allow CpU
exception in kernel partially") was broken.  The commit was to fix
theoretical problem but broke usual case.  Revert it for now.

Signed-off-by: Atsushi Nemoto 
Signed-off-by: Ralf Baechle

[MIPS] SMTC: Fix recursion in instant IPI replay code.

2007-03-29T22:46:36+00:00

local_irq_restore -> raw_local_irq_restore -> irq_restore_epilog ->
	smtc_ipi_replay -> smtc_ipi_dq -> spin_unlock_irqrestore ->
	_spin_unlock_irqrestore -> local_irq_restore

The recursion does abort when there is no more IPI queued for a CPU, so
this isn't usually fatal which is why we got away with this for so long
until this was discovered by code inspection.

Signed-off-by: Ralf Baechle

[MIPS] SMTC: Fix false trigger of debug code on single VPE.

2007-03-29T22:46:36+00:00

Make smtc_setup_irq() update the list of interrupts which need to be
watched by the debug code itself.  Also there is no need to initialize the
IPI swint when running with a single VPE, so don't initialize it.

Signed-off-by: Ralf Baechle

[MIPS] SMTC: irq_{enter,leave} and kstats keeping for relayed timer ints.

2007-03-29T22:46:35+00:00

Signed-off-by: Ralf Baechle

[MIPS] lockdep: Deal with interrupt disable hazard in TRACE_IRQFLAGS

2007-03-29T22:46:35+00:00

Between the mtc0 or di instruction that disables interrupts and the
following hazard barrier a processor may still take interrupts.  If an
interrupt is taken after interrupts are disabled but before the state
is updated it will appear to restore_all that it is incorrectly returning
with interrupts disabled.

Signed-off-by: Chris Dearman 
Signed-off-by: Ralf Baechle

[MIPS] lockdep: Handle interrupts in R3000 style c0_status register.

2007-03-29T22:46:35+00:00

Check the IEP bit for R3000 style processors when checking to see if
interrupts will be reenabled in restore_all.

Signed-off-by: Chris Dearman 
Signed-off-by: Ralf Baechle

[MIPS] Export except_vec_vi_{mori,lui,ori} as text symbols.

2007-03-19T20:22:44+00:00

Otherwise objdump will screw up disassembly.

Signed-off-by: Ralf Baechle

[MIPS] VI: TRACE_IRQS_OFF clobbers $v0, so save & restore around call.

2007-03-19T20:22:43+00:00

Signed-off-by: Ralf Baechle