diff options
| author |   Christophe Leroy <christophe.leroy@csgroup.eu> | 2021-10-19 09:29:17 +0200 |
|---|---|---|
| committer |   Michael Ellerman <mpe@ellerman.id.au> | 2021-12-09 22:41:17 +1100 |
| commit | 526d4a4c77aedf1b7df1133e5cced29c70232e6e (patch) | |
| tree | eae41ed69c6e5b71842206dcda4a9bc7e961bf51 /arch/powerpc/kernel/entry_32.S | |
| parent | powerpc/32s: Remove capability to disable KUEP at boottime (diff) | |
| download | linux-dev-526d4a4c77aedf1b7df1133e5cced29c70232e6e.tar.xz linux-dev-526d4a4c77aedf1b7df1133e5cced29c70232e6e.zip | |
powerpc/32s: Do kuep_lock() and kuep_unlock() in assembly
When interrupt and syscall entries where converted to C, KUEP locking
and unlocking was also converted. It improved performance by unrolling
the loop, and allowed easily implementing boot time deactivation of
KUEP.
However, null_syscall selftest shows that KUEP is still heavy
(361 cycles with KUEP, 212 cycles without).
A way to improve more is to group 'mtsr's together, instead of
repeating 'addi' + 'mtsr' several times.
In order to do that, more registers need to be available. In C, GCC
will always be able to provide the requested number of registers, but
at the cost of saving some data on the stack, which is counter
performant here.
So let's do it in assembly, when we have full control of which
register can be used. It also has the advantage of locking earlier
and unlocking later and it helps GCC generating less tricky code.
The only drawback is to make boot time deactivation less straight
forward and require 'hand' instruction patching.
Group 'mtsr's by 4.
With this change, null_syscall selftest reports 336 cycles. Without
the change it was 361 cycles, that's a 7% reduction.
Signed-off-by: Christophe Leroy <christophe.leroy@csgroup.eu>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/115cb279e9b9948dfd93a065e047081c59e3a2a6.1634627931.git.christophe.leroy@csgroup.eu
Diffstat (limited to 'arch/powerpc/kernel/entry_32.S')
| -rw-r--r-- | arch/powerpc/kernel/entry_32.S | 31 |
1 files changed, 31 insertions, 0 deletions
diff --git a/arch/powerpc/kernel/entry_32.S b/arch/powerpc/kernel/entry_32.S index c62dd9815965..0756829b2f7f 100644 --- a/arch/powerpc/kernel/entry_32.S +++ b/arch/powerpc/kernel/entry_32.S @@ -73,6 +73,34 @@ prepare_transfer_to_handler: _ASM_NOKPROBE_SYMBOL(prepare_transfer_to_handler) #endif /* CONFIG_PPC_BOOK3S_32 || CONFIG_E500 */ +#if defined(CONFIG_PPC_KUEP) && defined(CONFIG_PPC_BOOK3S_32) + .globl __kuep_lock +__kuep_lock: + mfsr r9,0 + rlwinm r9,r9,0,8,3 + oris r9,r9,SR_NX@h + update_user_segments_by_4 r9, r10, r11, r12 + blr + +__kuep_unlock: + mfsr r9,0 + rlwinm r9,r9,0,8,2 + update_user_segments_by_4 r9, r10, r11, r12 + blr + +.macro kuep_lock + bl __kuep_lock +.endm +.macro kuep_unlock + bl __kuep_unlock +.endm +#else +.macro kuep_lock +.endm +.macro kuep_unlock +.endm +#endif + .globl transfer_to_syscall transfer_to_syscall: stw r11, GPR1(r1) @@ -93,6 +121,7 @@ transfer_to_syscall: SAVE_GPRS(3, 8, r1) addi r2,r10,-THREAD SAVE_NVGPRS(r1) + kuep_lock /* Calling convention has r9 = orig r0, r10 = regs */ addi r10,r1,STACK_FRAME_OVERHEAD @@ -109,6 +138,7 @@ ret_from_syscall: cmplwi cr0,r5,0 bne- 2f #endif /* CONFIG_PPC_47x */ + kuep_unlock lwz r4,_LINK(r1) lwz r5,_CCR(r1) mtlr r4 @@ -272,6 +302,7 @@ interrupt_return: beq .Lkernel_interrupt_return bl interrupt_exit_user_prepare cmpwi r3,0 + kuep_unlock bne- .Lrestore_nvgprs .Lfast_user_interrupt_return: |