Commits · f30a5e68f026f3214a9392391537adaa79996b24 · Kirill Smelkov / linux

02 Jul, 2019 39 commits

powerpc/tm: update comment about interrupt re-entrancy · f30a5e68

Nicholas Piggin authored Jun 28, 2019

Since the system reset interrupt began to use its own stack, and
machine check interrupts have done so for some time, r1 can be
changed without clearing MSR[RI], provided no other interrupts
(including SLB misses) are taken.

MSR[RI] does have to be cleared when using SCRATCH0, however.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

f30a5e68

powerpc/64s/exception: move SET_SCRATCH0 into EXCEPTION_PROLOG_0 · d7fb34c7

Nicholas Piggin authored Jun 28, 2019

No generated code change.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

d7fb34c7

powerpc/64s/exception: denorm handler use standard scratch save macro · 904f81f3

Nicholas Piggin authored Jun 28, 2019

Although the 0x1500 interrupt only applies to bare metal, it is better
to just use the standard macro for scratch save.

Runtime code path remains unchanged (due to instruction patching).
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

904f81f3

powerpc/64s/exception: machine check use standard macros to save dar/dsisr · 02a12581
Nicholas Piggin authored Jun 28, 2019
```
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
```
02a12581

powerpc/64s/exception: add dar and dsisr options to exception macro · 5312c494

Nicholas Piggin authored Jun 28, 2019

Some exception entry requires DAR and/or DSISR to be saved into the
paca exception save area. Add options to the standard exception
macros for these.

Generated code changes slightly due to code structure.

-     554:      a6 02 72 7d     mfdsisr r11
-     558:      a8 00 4d f9     std     r10,168(r13)
-     55c:      b0 00 6d 91     stw     r11,176(r13)
+     554:      a8 00 4d f9     std     r10,168(r13)
+     558:      a6 02 52 7d     mfdsisr r10
+     55c:      b0 00 4d 91     stw     r10,176(r13)
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

5312c494

powerpc/64s/exception: use common macro for windup · 391e941b

Nicholas Piggin authored Jun 28, 2019

No generated code change.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

391e941b

powerpc/64s/exception: shuffle windup code around · b113c083

Nicholas Piggin authored Jun 28, 2019

Restore all SPRs and CR up-front, these are longer latency
instructions. Move register restore around to maximise pairs of
adjacent loads (e.g., restore r0 next to r1).
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

b113c083

powerpc/64s/exception: simplify hmi windup code · 67d4160a

Nicholas Piggin authored Jun 28, 2019

Duplicate the hmi windup code for both cases, rather than to put a
special case branch in the middle of it. Remove unused label. This
helps with later code consolidation.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

67d4160a

powerpc/64s/exception: move machine check windup in_mce handling · ad73d8d4

Nicholas Piggin authored Jun 28, 2019

Move in_mce decrement earlier before registers are restored (but
still after RI=0). This helps with later consolidation.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

ad73d8d4

powerpc/64s/exception: windup use r9 consistently to restore SPRs · 9592b29a

Nicholas Piggin authored Jun 28, 2019

Trivial code change, r3->r9.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

9592b29a

powerpc/64s/exception: mtmsrd L=1 cleanup · fbc50063

Nicholas Piggin authored Jun 28, 2019

All supported 64s CPUs support mtmsrd L=1 instruction, so a cleanup
can be made in sreset and mce handlers.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

fbc50063

powerpc/64s/exception: avoid SPR RAW scoreboard stall in real mode entry · 63d60d0c

Nicholas Piggin authored Jun 28, 2019

Move SPR reads ahead of writes. Real mode entry that is not a KVM
guest is rare these days, but bad practice propagates.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

63d60d0c

powerpc/64s/exception: clean up system call entry · b0b2a93d

Nicholas Piggin authored Jun 28, 2019

syscall / hcall entry unnecessarily differs between KVM and non-KVM
builds. Move the SMT priority instruction to the same location
(after INTERRUPT_TO_KERNEL).
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

b0b2a93d

powerpc/64s/exception: move paca save area offsets into exception-64s.S · 15820091

Nicholas Piggin authored Jun 22, 2019

No generated code change.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

15820091

powerpc/64s/exception: remove pointless EXCEPTION_PROLOG macro indirection · d064151f

Nicholas Piggin authored Jun 22, 2019

No generated code change. Final vmlinux is changed only due to change
in bug table line numbers.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

d064151f

powerpc/64s/exception: generate regs clear instructions using .rept · f3c8b6c6

Nicholas Piggin authored Jun 22, 2019

No generated code change.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

f3c8b6c6

powerpc/64s/exception: fix indenting irregularities · bf66e3c4

Nicholas Piggin authored Jun 22, 2019

Generally, macros that result in instructions being expanded are
indented by a tab, and those that don't have no indent. Fix the
obvious cases that go contrary to style.

No generated code change.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

bf66e3c4

powerpc/64s/exception: use a gas macro for system call handler code · 1b4d4a79

Nicholas Piggin authored Jun 22, 2019

No generated code change.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

1b4d4a79

powerpc/64s/exception: remove unused BRANCH_TO_COMMON · f945478d

Nicholas Piggin authored Jun 22, 2019

Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

f945478d

powerpc/64s/exception: remove __BRANCH_TO_KVM · 64e41351

Nicholas Piggin authored Jun 22, 2019

No generated code change.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

64e41351

powerpc/64s/exception: move head-64.h code to exception-64s.S where it is used · a0502434

Nicholas Piggin authored Jun 22, 2019

No generated code change.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

a0502434

powerpc/64s/exception: move exception-64s.h code to exception-64s.S where it is used · 12a04809
Nicholas Piggin authored Jun 22, 2019
```
No generated code change.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
```
12a04809

powerpc/64s/exception: move KVM related code together · f1ff37e8

Nicholas Piggin authored Jun 22, 2019

No generated code change.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

f1ff37e8

powerpc/64s/exception: remove STD_EXCEPTION_COMMON variants · 6d18f29c

Nicholas Piggin authored Jun 22, 2019

These are only called in one place each.

No generated code change.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

6d18f29c

powerpc/64s/exception: move EXCEPTION_PROLOG_2* to a more logical place · f0ac4478

Nicholas Piggin authored Jun 22, 2019

No generated code change.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

f0ac4478

powerpc/64s/exception: improve 0x500 handler code · 80bd9177

Nicholas Piggin authored Jun 22, 2019

After the previous cleanup, it becomes possible to consolidate some
common code outside the runtime alternate patching. Also remove
unused labels.

This results in some code change, but unchanged runtime instruction
sequence.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

80bd9177

powerpc/64s/exception: unwind exception-64s.h macros · fc557537

Nicholas Piggin authored Jun 22, 2019

Many of these macros just specify 1-4 lines which are only called a
few times each at most, and often just once. Remove this indirection.

No generated code change.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

fc557537

powerpc/64s/exception: Move EXCEPTION_COMMON additions into callers · 47169fba

Nicholas Piggin authored Jun 22, 2019

More cases of code insertion via macros that does not add a great
deal. All the additions have to be specified in the macro arguments,
so they can just as well go after the macro.

No generated code change.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

47169fba

powerpc/64s/exception: Move EXCEPTION_COMMON handler and return branches into callers · c06075f3

Nicholas Piggin authored Jun 22, 2019

The aim is to reduce the amount of indirection it takes to get through
the exception handler macros, particularly where it provides little
code sharing.

No generated code change.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

c06075f3

powerpc/64s/exception: Make EXCEPTION_PROLOG_0 a gas macro for consistency with others · 5dba1d50
Nicholas Piggin authored Jun 22, 2019
```
No generated code change.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
```
5dba1d50

powerpc/64s/exception: KVM handler can set the HSRR trap bit · c0c6cd15

Nicholas Piggin authored Jun 22, 2019

Move the KVM trap HSRR bit into the KVM handler, which can be
conditionally applied when hsrr parameter is set.

No generated code change.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

c0c6cd15

powerpc/64s/exception: merge KVM handler and skip variants · 17bdc064

Nicholas Piggin authored Jun 22, 2019

Conditionally expand the skip case if it is specified.

No generated code change.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

17bdc064

powerpc/64s/exception: consolidate maskable and non-maskable prologs · fa4cf6b7

Nicholas Piggin authored Jun 22, 2019

Conditionally expand the soft-masking test if a mask is passed in.

No generated code change.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

fa4cf6b7

powerpc/64s/exception: remove the "extra" macro parameter · a7c1ca19

Nicholas Piggin authored Jun 22, 2019

Rather than pass in the soft-masking and KVM tests via macro that is
passed to another macro to expand it, switch to usig gas macros and
conditionally expand the soft-masking and KVM tests.

The system reset with its idle test is open coded as it is a one-off.

No generated code change.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

a7c1ca19

powerpc/64s/exception: fix sreset KVM test code · 8f528359

Nicholas Piggin authored Jun 22, 2019

The sreset handler KVM test theoretically should not depend on P7.
In practice KVM now only supports P7 and up so no real bug fix, but
this change is made now so the quirk is not propagated through
cleanup patches.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

8f528359

powerpc/64s/exception: move and tidy EXCEPTION_PROLOG_2 variants · 2d046308

Nicholas Piggin authored Jun 22, 2019

- Re-name the macros to _REAL and _VIRT suffixes rather than no and
  _RELON suffix.

- Move the macro definitions together in the file.

- Move RELOCATABLE ifdef inside the _VIRT macro.

Further consolidation between variants does not buy much here.

No generated code change.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

2d046308

powerpc/64s/exception: consolidate EXCEPTION_PROLOG_2 with _NORI variant · bd7b6d13

Nicholas Piggin authored Jun 22, 2019

Switch to a gas macro that conditionally expands the RI clearing
instruction.

No generated code change.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

bd7b6d13

powerpc/64s/exception: remove H concatenation for EXC_HV variants · 4508a74a

Nicholas Piggin authored Jun 22, 2019

Replace all instances of this with gas macros that test the hsrr
parameter and use the appropriate register names / labels.

No generated code change.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
[mpe: Remove extraneous 2nd check for 0xea0 in SOFTEN_TEST]
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

4508a74a

powerpc/64s/exception: Remove unused SOFTEN_VALUE_0x980 · 1efd8caa

Michael Ellerman authored Jul 02, 2019

Remove SOFTEN_VALUE_0x980, it's been unused since commit
dabe859e ("powerpc: Give hypervisor decrementer interrupts their
own handler") (Sep 2012).
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

1efd8caa

01 Jul, 2019 1 commit

powerpc/64s/exception: fix line wrap and semicolon inconsistencies in macros · 4b1f5ccc

Nicholas Piggin authored Jun 12, 2019

By convention, all lines should be separated by a semicolons. Last line
should have neither semicolon or line wrap.

No generated code change.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

4b1f5ccc