From: Christophe Leroy <christophe.leroy@csgroup.eu>
To: Benjamin Herrenschmidt <benh@kernel.crashing.org>,
Paul Mackerras <paulus@samba.org>,
Michael Ellerman <mpe@ellerman.id.au>,
npiggin@gmail.com
Cc: linux-kernel@vger.kernel.org, linuxppc-dev@lists.ozlabs.org
Subject: [RFC PATCH v1 29/41] powerpc/32: Only restore non volatile registers when required
Date: Tue, 9 Feb 2021 09:56:54 +0000 (UTC) [thread overview]
Message-ID: <2eec585c9d60177a135924b5562dcf54a9463e05.1612864003.git.christophe.leroy@csgroup.eu> (raw)
In-Reply-To: <cover.1612864003.git.christophe.leroy@csgroup.eu>
Until now, non volatile registers were restored everytime they
were saved, ie using EXC_XFER_STD meant saving and restoring
them while EXC_XFER_LITE meant neither saving not restoring them.
Now that they are always saved, EXC_XFER_STD means to restore
them and EXC_XFER_LITE means to not restore them.
Most of the users of EXC_XFER_STD only need to retrieve the
non volatile registers. For them there is no need to restore
the non volatile registers as they have not been modified.
Only very few exceptions require non volatile registers restore.
Opencode the few places which require saving of non volatile
registers.
Signed-off-by: Christophe Leroy <christophe.leroy@csgroup.eu>
---
arch/powerpc/kernel/entry_32.S | 1 -
arch/powerpc/kernel/head_40x.S | 10 ++++++++--
arch/powerpc/kernel/head_8xx.S | 24 ++++++++++++++++++++----
arch/powerpc/kernel/head_book3s_32.S | 17 ++++++++++++++---
arch/powerpc/kernel/head_booke.h | 10 ++++++++--
arch/powerpc/kernel/head_fsl_booke.S | 16 ++++++++++++----
6 files changed, 62 insertions(+), 16 deletions(-)
diff --git a/arch/powerpc/kernel/entry_32.S b/arch/powerpc/kernel/entry_32.S
index 03f4d07f4176..d8a913f850ef 100644
--- a/arch/powerpc/kernel/entry_32.S
+++ b/arch/powerpc/kernel/entry_32.S
@@ -335,7 +335,6 @@ _ASM_NOKPROBE_SYMBOL(fast_exception_return)
.globl ret_from_except_full
ret_from_except_full:
- REST_NVGPRS(r1)
/* fall through */
.globl ret_from_except
diff --git a/arch/powerpc/kernel/head_40x.S b/arch/powerpc/kernel/head_40x.S
index 7270caff665c..f3e5b462113f 100644
--- a/arch/powerpc/kernel/head_40x.S
+++ b/arch/powerpc/kernel/head_40x.S
@@ -228,12 +228,18 @@ _ASM_NOKPROBE_SYMBOL(\name\()_virt)
/* 0x0600 - Alignment Exception */
START_EXCEPTION(0x0600, Alignment)
EXCEPTION_PROLOG 0x600 Alignment handle_dar_dsisr=1
- EXC_XFER_STD(0x600, alignment_exception)
+ prepare_transfer_to_handler
+ bl alignment_exception
+ REST_NVGPRS(r1)
+ b interrupt_return
/* 0x0700 - Program Exception */
START_EXCEPTION(0x0700, ProgramCheck)
EXCEPTION_PROLOG 0x700 ProgramCheck handle_dar_dsisr=1
- EXC_XFER_STD(0x700, program_check_exception)
+ prepare_transfer_to_handler
+ bl program_check_exception
+ REST_NVGPRS(r1)
+ b interrupt_return
EXCEPTION(0x0800, Trap_08, unknown_exception, EXC_XFER_STD)
EXCEPTION(0x0900, Trap_09, unknown_exception, EXC_XFER_STD)
diff --git a/arch/powerpc/kernel/head_8xx.S b/arch/powerpc/kernel/head_8xx.S
index c48de97f42fc..86f844eb0e5a 100644
--- a/arch/powerpc/kernel/head_8xx.S
+++ b/arch/powerpc/kernel/head_8xx.S
@@ -131,10 +131,18 @@ instruction_counter:
/* Alignment exception */
START_EXCEPTION(0x600, Alignment)
EXCEPTION_PROLOG 0x600 Alignment handle_dar_dsisr=1
- EXC_XFER_STD(0x600, alignment_exception)
+ prepare_transfer_to_handler
+ bl alignment_exception
+ REST_NVGPRS(r1)
+ b interrupt_return
/* Program check exception */
- EXCEPTION(0x700, ProgramCheck, program_check_exception, EXC_XFER_STD)
+ START_EXCEPTION(0x700, ProgramCheck)
+ EXCEPTION_PROLOG 0x700 ProgramCheck
+ prepare_transfer_to_handler
+ bl program_check_exception
+ REST_NVGPRS(r1)
+ b interrupt_return
/* Decrementer */
EXCEPTION(0x900, Decrementer, timer_interrupt, EXC_XFER_LITE)
@@ -149,7 +157,12 @@ instruction_counter:
/* On the MPC8xx, this is a software emulation interrupt. It occurs
* for all unimplemented and illegal instructions.
*/
- EXCEPTION(0x1000, SoftEmu, emulation_assist_interrupt, EXC_XFER_STD)
+ START_EXCEPTION(0x1000, SoftEmu)
+ EXCEPTION_PROLOG 0x1000 SoftEmu
+ prepare_transfer_to_handler
+ bl emulation_assist_interrupt
+ REST_NVGPRS(r1)
+ b interrupt_return
/*
* For the MPC8xx, this is a software tablewalk to load the instruction
@@ -348,7 +361,10 @@ DARFixed:/* Return from dcbx instruction bug workaround */
EXCEPTION_PROLOG_2 0x1c00 DataBreakpoint handle_dar_dsisr=1
mfspr r4,SPRN_BAR
stw r4,_DAR(r11)
- EXC_XFER_STD(0x1c00, do_break)
+ prepare_transfer_to_handler
+ bl do_break
+ REST_NVGPRS(r1)
+ b interrupt_return
#ifdef CONFIG_PERF_EVENTS
START_EXCEPTION(0x1d00, InstructionBreakpoint)
diff --git a/arch/powerpc/kernel/head_book3s_32.S b/arch/powerpc/kernel/head_book3s_32.S
index 67dac65b8ec3..609b2eedd4f9 100644
--- a/arch/powerpc/kernel/head_book3s_32.S
+++ b/arch/powerpc/kernel/head_book3s_32.S
@@ -300,7 +300,10 @@ ALT_MMU_FTR_SECTION_END_IFSET(MMU_FTR_HPTE_TABLE)
andis. r0, r5, DSISR_DABRMATCH@h
bne- 1f
EXC_XFER_LITE(0x300, do_page_fault)
-1: EXC_XFER_STD(0x300, do_break)
+1: prepare_transfer_to_handler
+ bl do_break
+ REST_NVGPRS(r1)
+ b interrupt_return
/* Instruction access exception. */
@@ -336,10 +339,18 @@ END_MMU_FTR_SECTION_IFSET(MMU_FTR_HPTE_TABLE)
/* Alignment exception */
START_EXCEPTION(0x600, Alignment)
EXCEPTION_PROLOG 0x600 Alignment handle_dar_dsisr=1
- EXC_XFER_STD(0x600, alignment_exception)
+ prepare_transfer_to_handler
+ bl alignment_exception
+ REST_NVGPRS(r1)
+ b interrupt_return
/* Program check exception */
- EXCEPTION(0x700, ProgramCheck, program_check_exception, EXC_XFER_STD)
+ START_EXCEPTION(0x700, ProgramCheck)
+ EXCEPTION_PROLOG 0x700 ProgramCheck
+ prepare_transfer_to_handler
+ bl program_check_exception
+ REST_NVGPRS(r1)
+ b interrupt_return
/* Floating-point unavailable */
START_EXCEPTION(0x800, FPUnavailable)
diff --git a/arch/powerpc/kernel/head_booke.h b/arch/powerpc/kernel/head_booke.h
index 9bc86aab65ef..809d1c6c9af9 100644
--- a/arch/powerpc/kernel/head_booke.h
+++ b/arch/powerpc/kernel/head_booke.h
@@ -483,14 +483,20 @@ ALT_FTR_SECTION_END_IFSET(CPU_FTR_EMB_HV)
NORMAL_EXCEPTION_PROLOG(0x600, ALIGNMENT); \
mfspr r4,SPRN_DEAR; /* Grab the DEAR and save it */ \
stw r4,_DEAR(r11); \
- EXC_XFER_STD(0x0600, alignment_exception)
+ prepare_transfer_to_handler; \
+ bl alignment_exception; \
+ REST_NVGPRS(r1); \
+ b interrupt_return
#define PROGRAM_EXCEPTION \
START_EXCEPTION(Program) \
NORMAL_EXCEPTION_PROLOG(0x700, PROGRAM); \
mfspr r4,SPRN_ESR; /* Grab the ESR and save it */ \
stw r4,_ESR(r11); \
- EXC_XFER_STD(0x0700, program_check_exception)
+ prepare_transfer_to_handler; \
+ bl program_check_exception; \
+ REST_NVGPRS(r1); \
+ b interrupt_return
#define DECREMENTER_EXCEPTION \
START_EXCEPTION(Decrementer) \
diff --git a/arch/powerpc/kernel/head_fsl_booke.S b/arch/powerpc/kernel/head_fsl_booke.S
index bf2730b4e43b..210871b2eb41 100644
--- a/arch/powerpc/kernel/head_fsl_booke.S
+++ b/arch/powerpc/kernel/head_fsl_booke.S
@@ -625,12 +625,20 @@ END_BTB_FLUSH_SECTION
/* SPE Floating Point Data */
#ifdef CONFIG_SPE
- EXCEPTION(0x2030, SPE_FP_DATA, SPEFloatingPointData,
- SPEFloatingPointException, EXC_XFER_STD)
+ START_EXCEPTION(SPEFloatingPointData)
+ NORMAL_EXCEPTION_PROLOG(0x2030, SPE_FP_DATA)
+ prepare_transfer_to_handler
+ bl SPEFloatingPointException
+ REST_NVGPRS(r1)
+ b interrupt_return
/* SPE Floating Point Round */
- EXCEPTION(0x2050, SPE_FP_ROUND, SPEFloatingPointRound, \
- SPEFloatingPointRoundException, EXC_XFER_STD)
+ START_EXCEPTION(SPEFloatingPointRound)
+ NORMAL_EXCEPTION_PROLOG(0x2050, SPE_FP_ROUND)
+ prepare_transfer_to_handler
+ bl SPEFloatingPointRoundException
+ REST_NVGPRS(r1)
+ b interrupt_return
#elif defined(CONFIG_SPE_POSSIBLE)
EXCEPTION(0x2040, SPE_FP_DATA, SPEFloatingPointData,
unknown_exception, EXC_XFER_STD)
--
2.25.0
next prev parent reply other threads:[~2021-02-09 10:12 UTC|newest]
Thread overview: 42+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-02-09 9:56 [RFC PATCH v1 00/41] powerpc/32: Switch to interrupt entry/exit in C Christophe Leroy
2021-02-09 9:56 ` [RFC PATCH v1 01/41] powerpc/32: Preserve cr1 in exception prolog stack check to fix build error Christophe Leroy
2021-02-09 9:56 ` [RFC PATCH v1 02/41] powerpc/40x: Don't use SPRN_SPRG_SCRATCH0/1 in TLB miss handlers Christophe Leroy
2021-02-09 9:56 ` [RFC PATCH v1 03/41] powerpc/40x: Change CRITICAL_EXCEPTION_PROLOG macro to a gas macro Christophe Leroy
2021-02-09 9:56 ` [RFC PATCH v1 04/41] powerpc/40x: Save SRR0/SRR1 and r10/r11 earlier in critical exception Christophe Leroy
2021-02-09 9:56 ` [RFC PATCH v1 05/41] powerpc/40x: Reorder a few instructions in critical exception prolog Christophe Leroy
2021-02-09 9:56 ` [RFC PATCH v1 06/41] powerpc/40x: Prepare for enabling MMU " Christophe Leroy
2021-02-09 9:56 ` [RFC PATCH v1 07/41] powerpc/40x: Prepare normal exception handler for enabling MMU early Christophe Leroy
2021-02-09 9:56 ` [RFC PATCH v1 08/41] powerpc/32: Reconcile interrupts in C Christophe Leroy
2021-02-09 9:56 ` [RFC PATCH v1 09/41] powerpc/32: Entry cpu time accounting " Christophe Leroy
2021-02-09 9:56 ` [RFC PATCH v1 10/41] powerpc/32: Handle bookE debugging in C in exception entry Christophe Leroy
2021-02-09 9:56 ` [RFC PATCH v1 11/41] powerpc/32: Use fast instruction to set MSR RI in exception prolog on 8xx Christophe Leroy
2021-02-09 9:56 ` [RFC PATCH v1 12/41] powerpc/32: Remove ksp_limit Christophe Leroy
2021-02-09 9:56 ` [RFC PATCH v1 13/41] powerpc/32: Always enable data translation in exception prolog Christophe Leroy
2021-02-09 9:56 ` [RFC PATCH v1 14/41] powerpc/32: Tag DAR in EXCEPTION_PROLOG_2 for the 8xx Christophe Leroy
2021-02-09 9:56 ` [RFC PATCH v1 15/41] powerpc/32: Enable instruction translation at the same time as data translation Christophe Leroy
2021-02-09 9:56 ` [RFC PATCH v1 16/41] powerpc/32: Statically initialise first emergency context Christophe Leroy
2021-02-09 9:56 ` [RFC PATCH v1 17/41] powerpc/32: Add vmap_stack_overflow label inside the macro Christophe Leroy
2021-02-09 9:56 ` [RFC PATCH v1 18/41] powerpc/32: Use START_EXCEPTION() as much as possible Christophe Leroy
2021-02-09 9:56 ` [RFC PATCH v1 19/41] powerpc/32: Move exception prolog code into .text once MMU is back on Christophe Leroy
2021-02-09 9:56 ` [RFC PATCH v1 20/41] powerpc/32: Provide a name to exception prolog continuation in virtual mode Christophe Leroy
2021-02-09 9:56 ` [RFC PATCH v1 21/41] powerpc/32: Refactor booke critical registers saving Christophe Leroy
2021-02-09 9:56 ` [RFC PATCH v1 22/41] powerpc/32: Perform normal function call in exception entry Christophe Leroy
2021-02-09 9:56 ` [RFC PATCH v1 23/41] powerpc/32: Always save non volatile registers on " Christophe Leroy
2021-02-09 9:56 ` [RFC PATCH v1 24/41] powerpc/32: Replace ASM exception exit by C exception exit from ppc64 Christophe Leroy
2021-02-09 9:56 ` [RFC PATCH v1 25/41] powerpc/32: Set regs parameter in r3 in transfer_to_handler Christophe Leroy
2021-02-09 9:56 ` [RFC PATCH v1 26/41] powerpc/32: Remove handle_page_fault() Christophe Leroy
2021-02-09 9:56 ` [RFC PATCH v1 27/41] powerpc/32: Save trap number on stack in exception prolog Christophe Leroy
2021-02-09 9:56 ` [RFC PATCH v1 28/41] powerpc/32: Add a prepare_transfer_to_handler macro for exception prologs Christophe Leroy
2021-02-09 9:56 ` Christophe Leroy [this message]
2021-02-09 9:56 ` [RFC PATCH v1 30/41] powerpc/32: Dismantle EXC_XFER_STD/LITE/TEMPLATE Christophe Leroy
2021-02-09 9:56 ` [RFC PATCH v1 31/41] powerpc/32: Remove the xfer parameter in EXCEPTION() macro Christophe Leroy
2021-02-09 9:56 ` [RFC PATCH v1 32/41] powerpc/32: Refactor saving of volatile registers in exception prologs Christophe Leroy
2021-02-09 9:56 ` [RFC PATCH v1 33/41] powerpc/32: Save remaining registers in exception prolog Christophe Leroy
2021-02-09 9:56 ` [RFC PATCH v1 34/41] powerpc/32: Set current->thread.regs in C interrupt entry Christophe Leroy
2021-02-09 9:57 ` [RFC PATCH v1 35/41] powerpc/32: Return directly from power_save_ppc32_restore() Christophe Leroy
2021-02-09 9:57 ` [RFC PATCH v1 36/41] powerpc/32: Only use prepare_transfer_to_handler function on book3s/32 and e500 Christophe Leroy
2021-02-09 9:57 ` [RFC PATCH v1 37/41] powerpc/32s: Move KUEP locking/unlocking in C Christophe Leroy
2021-02-09 9:57 ` [RFC PATCH v1 38/41] powerpc/64s: Make kuap_check_amr() and kuap_get_and_check_amr() generic Christophe Leroy
2021-02-09 9:57 ` [RFC PATCH v1 39/41] powerpc/32s: Create C version of kuap save/restore/check helpers Christophe Leroy
2021-02-09 9:57 ` [RFC PATCH v1 40/41] powerpc/8xx: " Christophe Leroy
2021-02-09 9:57 ` [RFC PATCH v1 41/41] powerpc/32: Manage KUAP in C Christophe Leroy
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=2eec585c9d60177a135924b5562dcf54a9463e05.1612864003.git.christophe.leroy@csgroup.eu \
--to=christophe.leroy@csgroup.eu \
--cc=benh@kernel.crashing.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linuxppc-dev@lists.ozlabs.org \
--cc=mpe@ellerman.id.au \
--cc=npiggin@gmail.com \
--cc=paulus@samba.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).