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: [PATCH v2 03/43] powerpc/40x: Don't use SPRN_SPRG_SCRATCH0/1 in TLB miss handlers Date: Tue, 9 Mar 2021 12:09:28 +0000 (UTC) [thread overview] Message-ID: <d09c387cf0d70bc6ca2a560ac572d9bbaa68793a.1615291471.git.christophe.leroy@csgroup.eu> (raw) In-Reply-To: <cover.1615291471.git.christophe.leroy@csgroup.eu> SPRN_SPRG_SCRATCH5 is used to save SPRN_PID. SPRN_SPRG_SCRATCH6 is already available. SPRN_PID is only 8 bits. We have r12 that contains CR. We only need to preserve CR0, so we have space available in r12 to save PID. Keep PID in r12 and free up SPRN_SPRG_SCRATCH5. Then In TLB miss handlers, instead of using SPRN_SPRG_SCRATCH0 and SPRN_SPRG_SCRATCH1, use SPRN_SPRG_SCRATCH5 and SPRN_SPRG_SCRATCH6 to avoid future conflicts with normal exception prologs. Signed-off-by: Christophe Leroy <christophe.leroy@csgroup.eu> --- arch/powerpc/kernel/head_40x.S | 39 ++++++++++++++++------------------ 1 file changed, 18 insertions(+), 21 deletions(-) diff --git a/arch/powerpc/kernel/head_40x.S b/arch/powerpc/kernel/head_40x.S index 24724a7dad49..383238a98f77 100644 --- a/arch/powerpc/kernel/head_40x.S +++ b/arch/powerpc/kernel/head_40x.S @@ -249,13 +249,13 @@ _ENTRY(saved_ksp_limit) * load TLB entries from the page table if they exist. */ START_EXCEPTION(0x1100, DTLBMiss) - mtspr SPRN_SPRG_SCRATCH0, r10 /* Save some working registers */ - mtspr SPRN_SPRG_SCRATCH1, r11 + mtspr SPRN_SPRG_SCRATCH5, r10 /* Save some working registers */ + mtspr SPRN_SPRG_SCRATCH6, r11 mtspr SPRN_SPRG_SCRATCH3, r12 mtspr SPRN_SPRG_SCRATCH4, r9 mfcr r12 mfspr r9, SPRN_PID - mtspr SPRN_SPRG_SCRATCH5, r9 + rlwimi r12, r9, 0, 0xff mfspr r10, SPRN_DEAR /* Get faulting address */ /* If we are faulting a kernel address, we have to use the @@ -316,13 +316,12 @@ _ENTRY(saved_ksp_limit) /* The bailout. Restore registers to pre-exception conditions * and call the heavyweights to help us out. */ - mfspr r9, SPRN_SPRG_SCRATCH5 - mtspr SPRN_PID, r9 - mtcr r12 + mtspr SPRN_PID, r12 + mtcrf 0x80, r12 mfspr r9, SPRN_SPRG_SCRATCH4 mfspr r12, SPRN_SPRG_SCRATCH3 - mfspr r11, SPRN_SPRG_SCRATCH1 - mfspr r10, SPRN_SPRG_SCRATCH0 + mfspr r11, SPRN_SPRG_SCRATCH6 + mfspr r10, SPRN_SPRG_SCRATCH5 b DataStorage /* 0x1200 - Instruction TLB Miss Exception @@ -330,13 +329,13 @@ _ENTRY(saved_ksp_limit) * registers and bailout to a different point. */ START_EXCEPTION(0x1200, ITLBMiss) - mtspr SPRN_SPRG_SCRATCH0, r10 /* Save some working registers */ - mtspr SPRN_SPRG_SCRATCH1, r11 + mtspr SPRN_SPRG_SCRATCH5, r10 /* Save some working registers */ + mtspr SPRN_SPRG_SCRATCH6, r11 mtspr SPRN_SPRG_SCRATCH3, r12 mtspr SPRN_SPRG_SCRATCH4, r9 mfcr r12 mfspr r9, SPRN_PID - mtspr SPRN_SPRG_SCRATCH5, r9 + rlwimi r12, r9, 0, 0xff mfspr r10, SPRN_SRR0 /* Get faulting address */ /* If we are faulting a kernel address, we have to use the @@ -397,13 +396,12 @@ _ENTRY(saved_ksp_limit) /* The bailout. Restore registers to pre-exception conditions * and call the heavyweights to help us out. */ - mfspr r9, SPRN_SPRG_SCRATCH5 - mtspr SPRN_PID, r9 - mtcr r12 + mtspr SPRN_PID, r12 + mtcrf 0x80, r12 mfspr r9, SPRN_SPRG_SCRATCH4 mfspr r12, SPRN_SPRG_SCRATCH3 - mfspr r11, SPRN_SPRG_SCRATCH1 - mfspr r10, SPRN_SPRG_SCRATCH0 + mfspr r11, SPRN_SPRG_SCRATCH6 + mfspr r10, SPRN_SPRG_SCRATCH5 b InstructionAccess EXCEPTION(0x1300, Trap_13, unknown_exception, EXC_XFER_STD) @@ -543,13 +541,12 @@ finish_tlb_load: /* Done...restore registers and get out of here. */ - mfspr r9, SPRN_SPRG_SCRATCH5 - mtspr SPRN_PID, r9 - mtcr r12 + mtspr SPRN_PID, r12 + mtcrf 0x80, r12 mfspr r9, SPRN_SPRG_SCRATCH4 mfspr r12, SPRN_SPRG_SCRATCH3 - mfspr r11, SPRN_SPRG_SCRATCH1 - mfspr r10, SPRN_SPRG_SCRATCH0 + mfspr r11, SPRN_SPRG_SCRATCH6 + mfspr r10, SPRN_SPRG_SCRATCH5 rfi /* Should sync shadow TLBs */ b . /* prevent prefetch past rfi */ -- 2.25.0
WARNING: multiple messages have this Message-ID (diff)
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: linuxppc-dev@lists.ozlabs.org, linux-kernel@vger.kernel.org Subject: [PATCH v2 03/43] powerpc/40x: Don't use SPRN_SPRG_SCRATCH0/1 in TLB miss handlers Date: Tue, 9 Mar 2021 12:09:28 +0000 (UTC) [thread overview] Message-ID: <d09c387cf0d70bc6ca2a560ac572d9bbaa68793a.1615291471.git.christophe.leroy@csgroup.eu> (raw) In-Reply-To: <cover.1615291471.git.christophe.leroy@csgroup.eu> SPRN_SPRG_SCRATCH5 is used to save SPRN_PID. SPRN_SPRG_SCRATCH6 is already available. SPRN_PID is only 8 bits. We have r12 that contains CR. We only need to preserve CR0, so we have space available in r12 to save PID. Keep PID in r12 and free up SPRN_SPRG_SCRATCH5. Then In TLB miss handlers, instead of using SPRN_SPRG_SCRATCH0 and SPRN_SPRG_SCRATCH1, use SPRN_SPRG_SCRATCH5 and SPRN_SPRG_SCRATCH6 to avoid future conflicts with normal exception prologs. Signed-off-by: Christophe Leroy <christophe.leroy@csgroup.eu> --- arch/powerpc/kernel/head_40x.S | 39 ++++++++++++++++------------------ 1 file changed, 18 insertions(+), 21 deletions(-) diff --git a/arch/powerpc/kernel/head_40x.S b/arch/powerpc/kernel/head_40x.S index 24724a7dad49..383238a98f77 100644 --- a/arch/powerpc/kernel/head_40x.S +++ b/arch/powerpc/kernel/head_40x.S @@ -249,13 +249,13 @@ _ENTRY(saved_ksp_limit) * load TLB entries from the page table if they exist. */ START_EXCEPTION(0x1100, DTLBMiss) - mtspr SPRN_SPRG_SCRATCH0, r10 /* Save some working registers */ - mtspr SPRN_SPRG_SCRATCH1, r11 + mtspr SPRN_SPRG_SCRATCH5, r10 /* Save some working registers */ + mtspr SPRN_SPRG_SCRATCH6, r11 mtspr SPRN_SPRG_SCRATCH3, r12 mtspr SPRN_SPRG_SCRATCH4, r9 mfcr r12 mfspr r9, SPRN_PID - mtspr SPRN_SPRG_SCRATCH5, r9 + rlwimi r12, r9, 0, 0xff mfspr r10, SPRN_DEAR /* Get faulting address */ /* If we are faulting a kernel address, we have to use the @@ -316,13 +316,12 @@ _ENTRY(saved_ksp_limit) /* The bailout. Restore registers to pre-exception conditions * and call the heavyweights to help us out. */ - mfspr r9, SPRN_SPRG_SCRATCH5 - mtspr SPRN_PID, r9 - mtcr r12 + mtspr SPRN_PID, r12 + mtcrf 0x80, r12 mfspr r9, SPRN_SPRG_SCRATCH4 mfspr r12, SPRN_SPRG_SCRATCH3 - mfspr r11, SPRN_SPRG_SCRATCH1 - mfspr r10, SPRN_SPRG_SCRATCH0 + mfspr r11, SPRN_SPRG_SCRATCH6 + mfspr r10, SPRN_SPRG_SCRATCH5 b DataStorage /* 0x1200 - Instruction TLB Miss Exception @@ -330,13 +329,13 @@ _ENTRY(saved_ksp_limit) * registers and bailout to a different point. */ START_EXCEPTION(0x1200, ITLBMiss) - mtspr SPRN_SPRG_SCRATCH0, r10 /* Save some working registers */ - mtspr SPRN_SPRG_SCRATCH1, r11 + mtspr SPRN_SPRG_SCRATCH5, r10 /* Save some working registers */ + mtspr SPRN_SPRG_SCRATCH6, r11 mtspr SPRN_SPRG_SCRATCH3, r12 mtspr SPRN_SPRG_SCRATCH4, r9 mfcr r12 mfspr r9, SPRN_PID - mtspr SPRN_SPRG_SCRATCH5, r9 + rlwimi r12, r9, 0, 0xff mfspr r10, SPRN_SRR0 /* Get faulting address */ /* If we are faulting a kernel address, we have to use the @@ -397,13 +396,12 @@ _ENTRY(saved_ksp_limit) /* The bailout. Restore registers to pre-exception conditions * and call the heavyweights to help us out. */ - mfspr r9, SPRN_SPRG_SCRATCH5 - mtspr SPRN_PID, r9 - mtcr r12 + mtspr SPRN_PID, r12 + mtcrf 0x80, r12 mfspr r9, SPRN_SPRG_SCRATCH4 mfspr r12, SPRN_SPRG_SCRATCH3 - mfspr r11, SPRN_SPRG_SCRATCH1 - mfspr r10, SPRN_SPRG_SCRATCH0 + mfspr r11, SPRN_SPRG_SCRATCH6 + mfspr r10, SPRN_SPRG_SCRATCH5 b InstructionAccess EXCEPTION(0x1300, Trap_13, unknown_exception, EXC_XFER_STD) @@ -543,13 +541,12 @@ finish_tlb_load: /* Done...restore registers and get out of here. */ - mfspr r9, SPRN_SPRG_SCRATCH5 - mtspr SPRN_PID, r9 - mtcr r12 + mtspr SPRN_PID, r12 + mtcrf 0x80, r12 mfspr r9, SPRN_SPRG_SCRATCH4 mfspr r12, SPRN_SPRG_SCRATCH3 - mfspr r11, SPRN_SPRG_SCRATCH1 - mfspr r10, SPRN_SPRG_SCRATCH0 + mfspr r11, SPRN_SPRG_SCRATCH6 + mfspr r10, SPRN_SPRG_SCRATCH5 rfi /* Should sync shadow TLBs */ b . /* prevent prefetch past rfi */ -- 2.25.0
next prev parent reply other threads:[~2021-03-09 12:10 UTC|newest] Thread overview: 120+ messages / expand[flat|nested] mbox.gz Atom feed top 2021-03-09 12:09 [PATCH v2 00/43] powerpc/32: Switch to interrupt entry/exit in C Christophe Leroy 2021-03-09 12:09 ` Christophe Leroy 2021-03-09 12:09 ` [PATCH v2 01/43] powerpc/traps: unrecoverable_exception() is not an interrupt handler Christophe Leroy 2021-03-09 12:09 ` Christophe Leroy 2021-03-10 1:16 ` Nicholas Piggin 2021-03-10 1:16 ` Nicholas Piggin 2021-03-09 12:09 ` [PATCH v2 02/43] powerpc/traps: Declare unrecoverable_exception() as __noreturn Christophe Leroy 2021-03-09 12:09 ` Christophe Leroy 2021-03-10 1:22 ` Nicholas Piggin 2021-03-10 1:22 ` Nicholas Piggin 2021-03-12 8:40 ` Christophe Leroy 2021-03-12 8:40 ` Christophe Leroy 2021-03-09 12:09 ` Christophe Leroy [this message] 2021-03-09 12:09 ` [PATCH v2 03/43] powerpc/40x: Don't use SPRN_SPRG_SCRATCH0/1 in TLB miss handlers Christophe Leroy 2021-03-09 12:09 ` [PATCH v2 04/43] powerpc/40x: Change CRITICAL_EXCEPTION_PROLOG macro to a gas macro Christophe Leroy 2021-03-09 12:09 ` Christophe Leroy 2021-03-09 12:09 ` [PATCH v2 05/43] powerpc/40x: Save SRR0/SRR1 and r10/r11 earlier in critical exception Christophe Leroy 2021-03-09 12:09 ` Christophe Leroy 2021-03-09 12:09 ` [PATCH v2 06/43] powerpc/40x: Reorder a few instructions in critical exception prolog Christophe Leroy 2021-03-09 12:09 ` Christophe Leroy 2021-03-09 12:09 ` [PATCH v2 07/43] powerpc/40x: Prepare for enabling MMU " Christophe Leroy 2021-03-09 12:09 ` Christophe Leroy 2021-03-09 12:09 ` [PATCH v2 08/43] powerpc/40x: Prepare normal exception handler for enabling MMU early Christophe Leroy 2021-03-09 12:09 ` Christophe Leroy 2021-03-09 12:09 ` [PATCH v2 09/43] powerpc/32: Reconcile interrupts in C Christophe Leroy 2021-03-09 12:09 ` Christophe Leroy 2021-03-09 12:09 ` [PATCH v2 10/43] powerpc/32: Entry cpu time accounting " Christophe Leroy 2021-03-09 12:09 ` Christophe Leroy 2021-03-09 12:09 ` [PATCH v2 11/43] powerpc/32: Handle bookE debugging in C in exception entry Christophe Leroy 2021-03-09 12:09 ` Christophe Leroy 2021-03-09 12:09 ` [PATCH v2 12/43] powerpc/32: Use fast instruction to set MSR RI in exception prolog on 8xx Christophe Leroy 2021-03-09 12:09 ` Christophe Leroy 2021-03-09 12:09 ` [PATCH v2 13/43] powerpc/32: Remove ksp_limit Christophe Leroy 2021-03-09 12:09 ` Christophe Leroy 2021-03-09 12:09 ` [PATCH v2 14/43] powerpc/32: Always enable data translation in exception prolog Christophe Leroy 2021-03-09 12:09 ` Christophe Leroy 2021-03-09 12:09 ` [PATCH v2 15/43] powerpc/32: Tag DAR in EXCEPTION_PROLOG_2 for the 8xx Christophe Leroy 2021-03-09 12:09 ` Christophe Leroy 2021-03-09 12:09 ` [PATCH v2 16/43] powerpc/32: Enable instruction translation at the same time as data translation Christophe Leroy 2021-03-09 12:09 ` Christophe Leroy 2021-03-09 12:09 ` [PATCH v2 17/43] powerpc/32: Statically initialise first emergency context Christophe Leroy 2021-03-09 12:09 ` Christophe Leroy 2021-03-09 12:09 ` [PATCH v2 18/43] powerpc/32: Add vmap_stack_overflow label inside the macro Christophe Leroy 2021-03-09 12:09 ` Christophe Leroy 2021-03-09 12:09 ` [PATCH v2 19/43] powerpc/32: Use START_EXCEPTION() as much as possible Christophe Leroy 2021-03-09 12:09 ` Christophe Leroy 2021-03-09 12:09 ` [PATCH v2 20/43] powerpc/32: Move exception prolog code into .text once MMU is back on Christophe Leroy 2021-03-09 12:09 ` Christophe Leroy 2021-03-09 12:09 ` [PATCH v2 21/43] powerpc/32: Provide a name to exception prolog continuation in virtual mode Christophe Leroy 2021-03-09 12:09 ` Christophe Leroy 2021-03-09 12:09 ` [PATCH v2 22/43] powerpc/32: Refactor booke critical registers saving Christophe Leroy 2021-03-09 12:09 ` Christophe Leroy 2021-03-09 12:09 ` [PATCH v2 23/43] powerpc/32: Perform normal function call in exception entry Christophe Leroy 2021-03-09 12:09 ` Christophe Leroy 2021-03-09 12:09 ` [PATCH v2 24/43] powerpc/32: Always save non volatile registers on " Christophe Leroy 2021-03-09 12:09 ` Christophe Leroy 2021-03-09 12:09 ` [PATCH v2 25/43] powerpc/32: Replace ASM exception exit by C exception exit from ppc64 Christophe Leroy 2021-03-09 12:09 ` Christophe Leroy 2021-03-11 13:46 ` Michael Ellerman 2021-03-11 13:46 ` Michael Ellerman 2021-03-11 19:39 ` Christophe Leroy 2021-03-11 19:39 ` Christophe Leroy 2021-03-11 23:26 ` Michael Ellerman 2021-03-11 23:26 ` Michael Ellerman 2021-03-12 8:28 ` Christophe Leroy 2021-03-12 8:28 ` Christophe Leroy 2021-03-09 12:09 ` [PATCH v2 26/43] powerpc/32: Set regs parameter in r3 in transfer_to_handler Christophe Leroy 2021-03-09 12:09 ` Christophe Leroy 2021-03-09 12:09 ` [PATCH v2 27/43] powerpc/32: Call bad_page_fault() from do_page_fault() Christophe Leroy 2021-03-09 12:09 ` Christophe Leroy 2021-03-09 12:09 ` [PATCH v2 28/43] powerpc/64e: " Christophe Leroy 2021-03-09 12:09 ` Christophe Leroy 2021-03-10 1:29 ` Nicholas Piggin 2021-03-10 1:29 ` Nicholas Piggin 2021-03-12 8:39 ` Christophe Leroy 2021-03-12 8:39 ` Christophe Leroy 2021-03-09 12:09 ` [PATCH v2 29/43] powerpc/32: Save trap number on stack in exception prolog Christophe Leroy 2021-03-09 12:09 ` Christophe Leroy 2021-03-09 12:09 ` [PATCH v2 30/43] powerpc/32: Add a prepare_transfer_to_handler macro for exception prologs Christophe Leroy 2021-03-09 12:09 ` Christophe Leroy 2021-03-09 12:09 ` [PATCH v2 31/43] powerpc/32: Only restore non volatile registers when required Christophe Leroy 2021-03-09 12:09 ` Christophe Leroy 2021-03-09 12:09 ` [PATCH v2 32/43] powerpc/32: Dismantle EXC_XFER_STD/LITE/TEMPLATE Christophe Leroy 2021-03-09 12:09 ` Christophe Leroy 2021-03-09 12:09 ` [PATCH v2 33/43] powerpc/32: Remove the xfer parameter in EXCEPTION() macro Christophe Leroy 2021-03-09 12:09 ` Christophe Leroy 2021-03-09 12:10 ` [PATCH v2 34/43] powerpc/32: Refactor saving of volatile registers in exception prologs Christophe Leroy 2021-03-09 12:10 ` Christophe Leroy 2021-03-09 12:10 ` [PATCH v2 35/43] powerpc/32: Save remaining registers in exception prolog Christophe Leroy 2021-03-09 12:10 ` Christophe Leroy 2021-03-09 12:10 ` [PATCH v2 36/43] powerpc/32: Set current->thread.regs in C interrupt entry Christophe Leroy 2021-03-09 12:10 ` Christophe Leroy 2021-03-10 1:33 ` Nicholas Piggin 2021-03-10 1:33 ` Nicholas Piggin 2021-03-11 10:38 ` Christophe Leroy 2021-03-11 10:38 ` Christophe Leroy 2021-03-11 12:38 ` Christophe Leroy 2021-03-11 12:38 ` Christophe Leroy 2021-03-12 1:00 ` Nicholas Piggin 2021-03-12 1:00 ` Nicholas Piggin 2021-03-09 12:10 ` [PATCH v2 37/43] powerpc/32: Return directly from power_save_ppc32_restore() Christophe Leroy 2021-03-09 12:10 ` Christophe Leroy 2021-03-09 12:10 ` [PATCH v2 38/43] powerpc/32: Only use prepare_transfer_to_handler function on book3s/32 and e500 Christophe Leroy 2021-03-09 12:10 ` Christophe Leroy 2021-03-09 12:10 ` [PATCH v2 39/43] powerpc/32s: Move KUEP locking/unlocking in C Christophe Leroy 2021-03-09 12:10 ` Christophe Leroy 2021-03-09 12:10 ` [PATCH v2 40/43] powerpc/64s: Make kuap_check_amr() and kuap_get_and_check_amr() generic Christophe Leroy 2021-03-09 12:10 ` Christophe Leroy 2021-03-10 1:37 ` Nicholas Piggin 2021-03-10 1:37 ` Nicholas Piggin 2021-03-12 8:37 ` Christophe Leroy 2021-03-12 8:37 ` Christophe Leroy 2021-03-09 12:10 ` [PATCH v2 41/43] powerpc/32s: Create C version of kuap save/restore/check helpers Christophe Leroy 2021-03-09 12:10 ` Christophe Leroy 2021-03-09 12:10 ` [PATCH v2 42/43] powerpc/8xx: " Christophe Leroy 2021-03-09 12:10 ` Christophe Leroy 2021-03-09 12:10 ` [PATCH v2 43/43] powerpc/32: Manage KUAP in C Christophe Leroy 2021-03-09 12:10 ` Christophe Leroy 2021-03-14 10:01 ` [PATCH v2 00/43] powerpc/32: Switch to interrupt entry/exit " Michael Ellerman 2021-03-14 10:01 ` Michael Ellerman
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=d09c387cf0d70bc6ca2a560ac572d9bbaa68793a.1615291471.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: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.