From: "Christophe Leroy (CS GROUP)" <chleroy@kernel.org>
To: Mukesh Kumar Chaurasiya <mkchauras@linux.ibm.com>,
maddy@linux.ibm.com, mpe@ellerman.id.au, npiggin@gmail.com,
oleg@redhat.com, kees@kernel.org, luto@amacapital.net,
wad@chromium.org, mchauras@linux.ibm.com, thuth@redhat.com,
sshegde@linux.ibm.com, charlie@rivosinc.com, macro@orcam.me.uk,
akpm@linux-foundation.org, ldv@strace.io, deller@gmx.de,
ankur.a.arora@oracle.com, segher@kernel.crashing.org,
tglx@linutronix.de, thomas.weissschuh@linutronix.de,
peterz@infradead.org, menglong8.dong@gmail.com,
bigeasy@linutronix.de, namcao@linutronix.de,
kan.liang@linux.intel.com, mingo@kernel.org,
atrajeev@linux.vnet.ibm.com, mark.barnett@arm.com,
linuxppc-dev@lists.ozlabs.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH v2 5/8] powerpc: add exit_flags field in pt_regs
Date: Tue, 16 Dec 2025 10:52:42 +0100 [thread overview]
Message-ID: <dc0edfb4-d2bc-43be-9221-a1acb1cae5ec@kernel.org> (raw)
In-Reply-To: <20251214130245.43664-6-mkchauras@linux.ibm.com>
Le 14/12/2025 à 14:02, Mukesh Kumar Chaurasiya a écrit :
> From: Mukesh Kumar Chaurasiya <mchauras@linux.ibm.com>
>
> Add a new field `exit_flags` in the pt_regs structure. This field will hold
> the flags set during interrupt or syscall execution that are required during
> exit to user mode.
>
> Specifically, the `TIF_RESTOREALL` flag, stored in this field, helps the
> exit routine determine if any NVGPRs were modified and need to be restored
> before returning to userspace.
In the current implementation we did our best to keep this information
in a local var for performance reasons. Have you assessed the
performance impact of going through the stack for that ?
>
> This addition ensures a clean and architecture-specific mechanism to track
> per-syscall or per-interrupt state transitions related to register restore.
>
> Changes:
> - Add `exit_flags` and `__pt_regs_pad` to maintain 16-byte stack alignment
> - Update asm-offsets.c and ptrace.c for offset and validation
> - Update PT_* constants in uapi header to reflect the new layout
>
> Signed-off-by: Mukesh Kumar Chaurasiya <mchauras@linux.ibm.com>
> ---
> arch/powerpc/include/asm/ptrace.h | 3 +++
> arch/powerpc/include/uapi/asm/ptrace.h | 14 +++++++++-----
> arch/powerpc/kernel/asm-offsets.c | 1 +
> arch/powerpc/kernel/ptrace/ptrace.c | 1 +
> 4 files changed, 14 insertions(+), 5 deletions(-)
>
> diff --git a/arch/powerpc/include/asm/ptrace.h b/arch/powerpc/include/asm/ptrace.h
> index 94aa1de2b06e..3af8a5898fe3 100644
> --- a/arch/powerpc/include/asm/ptrace.h
> +++ b/arch/powerpc/include/asm/ptrace.h
> @@ -53,6 +53,9 @@ struct pt_regs
> unsigned long esr;
> };
> unsigned long result;
> + unsigned long exit_flags;
> + /* Maintain 16 byte interrupt stack alignment */
On powerpc/32, one 'long' is 4 bytes not 8.
> + unsigned long __pt_regs_pad[1];
> };
> };
> #if defined(CONFIG_PPC64) || defined(CONFIG_PPC_KUAP)
> diff --git a/arch/powerpc/include/uapi/asm/ptrace.h b/arch/powerpc/include/uapi/asm/ptrace.h
> index 01e630149d48..de56b216c9c5 100644
> --- a/arch/powerpc/include/uapi/asm/ptrace.h
> +++ b/arch/powerpc/include/uapi/asm/ptrace.h
> @@ -55,6 +55,8 @@ struct pt_regs
> unsigned long dar; /* Fault registers */
> unsigned long dsisr; /* on 4xx/Book-E used for ESR */
> unsigned long result; /* Result of a system call */
> + unsigned long exit_flags; /* System call exit flags */
> + unsigned long __pt_regs_pad[1]; /* Maintain 16 byte interrupt stack alignment */
On powerpc/32, one 'long' is 4 bytes not 8.
> };
>
> #endif /* __ASSEMBLER__ */
> @@ -114,10 +116,12 @@ struct pt_regs
> #define PT_DAR 41
> #define PT_DSISR 42
> #define PT_RESULT 43
> -#define PT_DSCR 44
> -#define PT_REGS_COUNT 44
> +#define PT_EXIT_FLAGS 44
> +#define PT_PAD 45
> +#define PT_DSCR 46
> +#define PT_REGS_COUNT 46
>
> -#define PT_FPR0 48 /* each FP reg occupies 2 slots in this space */
> +#define PT_FPR0 (PT_REGS_COUNT + 4) /* each FP reg occupies 2 slots in this space */
>
> #ifndef __powerpc64__
>
> @@ -129,7 +133,7 @@ struct pt_regs
> #define PT_FPSCR (PT_FPR0 + 32) /* each FP reg occupies 1 slot in 64-bit space */
>
>
> -#define PT_VR0 82 /* each Vector reg occupies 2 slots in 64-bit */
> +#define PT_VR0 (PT_FPSCR + 2) /* <82> each Vector reg occupies 2 slots in 64-bit */
> #define PT_VSCR (PT_VR0 + 32*2 + 1)
> #define PT_VRSAVE (PT_VR0 + 33*2)
>
> @@ -137,7 +141,7 @@ struct pt_regs
> /*
> * Only store first 32 VSRs here. The second 32 VSRs in VR0-31
> */
> -#define PT_VSR0 150 /* each VSR reg occupies 2 slots in 64-bit */
> +#define PT_VSR0 (PT_VRSAVE + 2) /* each VSR reg occupies 2 slots in 64-bit */
> #define PT_VSR31 (PT_VSR0 + 2*31)
> #endif /* __powerpc64__ */
>
> diff --git a/arch/powerpc/kernel/asm-offsets.c b/arch/powerpc/kernel/asm-offsets.c
> index a4bc80b30410..c0bb09f1db78 100644
> --- a/arch/powerpc/kernel/asm-offsets.c
> +++ b/arch/powerpc/kernel/asm-offsets.c
> @@ -292,6 +292,7 @@ int main(void)
> STACK_PT_REGS_OFFSET(_ESR, esr);
> STACK_PT_REGS_OFFSET(ORIG_GPR3, orig_gpr3);
> STACK_PT_REGS_OFFSET(RESULT, result);
> + STACK_PT_REGS_OFFSET(EXIT_FLAGS, exit_flags);
Where is that used ?
> STACK_PT_REGS_OFFSET(_TRAP, trap);
> #ifdef CONFIG_PPC64
> STACK_PT_REGS_OFFSET(SOFTE, softe);
> diff --git a/arch/powerpc/kernel/ptrace/ptrace.c b/arch/powerpc/kernel/ptrace/ptrace.c
> index c6997df63287..2134b6d155ff 100644
> --- a/arch/powerpc/kernel/ptrace/ptrace.c
> +++ b/arch/powerpc/kernel/ptrace/ptrace.c
> @@ -432,6 +432,7 @@ void __init pt_regs_check(void)
> CHECK_REG(PT_DAR, dar);
> CHECK_REG(PT_DSISR, dsisr);
> CHECK_REG(PT_RESULT, result);
> + CHECK_REG(PT_EXIT_FLAGS, exit_flags);
> #undef CHECK_REG
>
> BUILD_BUG_ON(PT_REGS_COUNT != sizeof(struct user_pt_regs) / sizeof(unsigned long));
next prev parent reply other threads:[~2025-12-16 9:53 UTC|newest]
Thread overview: 37+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-12-14 13:02 [PATCH v2 0/8] Generic IRQ entry/exit support for powerpc Mukesh Kumar Chaurasiya
2025-12-14 13:02 ` [PATCH v2 1/8] powerpc: rename arch_irq_disabled_regs Mukesh Kumar Chaurasiya
2025-12-14 13:02 ` [PATCH v2 2/8] powerpc: Prepare to build with generic entry/exit framework Mukesh Kumar Chaurasiya
2025-12-16 9:27 ` Christophe Leroy (CS GROUP)
2025-12-16 14:42 ` Mukesh Kumar Chaurasiya
2025-12-14 13:02 ` [PATCH v2 3/8] powerpc: introduce arch_enter_from_user_mode Mukesh Kumar Chaurasiya
2025-12-16 9:38 ` Christophe Leroy (CS GROUP)
2025-12-16 14:47 ` Mukesh Kumar Chaurasiya
2025-12-14 13:02 ` [PATCH v2 4/8] powerpc: Introduce syscall exit arch functions Mukesh Kumar Chaurasiya
2025-12-16 9:46 ` Christophe Leroy (CS GROUP)
2025-12-16 14:51 ` Mukesh Kumar Chaurasiya
2025-12-14 13:02 ` [PATCH v2 5/8] powerpc: add exit_flags field in pt_regs Mukesh Kumar Chaurasiya
2025-12-16 9:52 ` Christophe Leroy (CS GROUP) [this message]
2025-12-16 14:56 ` Mukesh Kumar Chaurasiya
2025-12-14 13:02 ` [PATCH v2 6/8] powerpc: Prepare for IRQ entry exit Mukesh Kumar Chaurasiya
2025-12-16 9:58 ` Christophe Leroy (CS GROUP)
2025-12-16 15:00 ` Mukesh Kumar Chaurasiya
2025-12-16 22:40 ` Christophe Leroy (CS GROUP)
2025-12-17 4:43 ` Mukesh Kumar Chaurasiya
2025-12-19 4:56 ` Mukesh Kumar Chaurasiya
2025-12-14 13:02 ` [PATCH v2 7/8] powerpc: Enable IRQ generic entry/exit path Mukesh Kumar Chaurasiya
2025-12-16 6:29 ` kernel test robot
2025-12-16 15:02 ` Mukesh Kumar Chaurasiya
2025-12-16 10:43 ` Christophe Leroy (CS GROUP)
2025-12-16 15:06 ` Mukesh Kumar Chaurasiya
2025-12-17 2:10 ` kernel test robot
2025-12-17 21:32 ` kernel test robot
2025-12-14 13:02 ` [PATCH v2 8/8] powerpc: Enable Generic Entry/Exit for syscalls Mukesh Kumar Chaurasiya
2025-12-14 16:20 ` Segher Boessenkool
2025-12-15 18:32 ` Mukesh Kumar Chaurasiya
2025-12-15 20:27 ` kernel test robot
2025-12-16 15:08 ` Mukesh Kumar Chaurasiya
2025-12-16 22:57 ` Christophe Leroy (CS GROUP)
2025-12-16 6:41 ` Christophe Leroy (CS GROUP)
2025-12-16 15:09 ` Mukesh Kumar Chaurasiya
2025-12-16 11:01 ` Christophe Leroy (CS GROUP)
2025-12-16 15:13 ` Mukesh Kumar Chaurasiya
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=dc0edfb4-d2bc-43be-9221-a1acb1cae5ec@kernel.org \
--to=chleroy@kernel.org \
--cc=akpm@linux-foundation.org \
--cc=ankur.a.arora@oracle.com \
--cc=atrajeev@linux.vnet.ibm.com \
--cc=bigeasy@linutronix.de \
--cc=charlie@rivosinc.com \
--cc=deller@gmx.de \
--cc=kan.liang@linux.intel.com \
--cc=kees@kernel.org \
--cc=ldv@strace.io \
--cc=linux-kernel@vger.kernel.org \
--cc=linuxppc-dev@lists.ozlabs.org \
--cc=luto@amacapital.net \
--cc=macro@orcam.me.uk \
--cc=maddy@linux.ibm.com \
--cc=mark.barnett@arm.com \
--cc=mchauras@linux.ibm.com \
--cc=menglong8.dong@gmail.com \
--cc=mingo@kernel.org \
--cc=mkchauras@linux.ibm.com \
--cc=mpe@ellerman.id.au \
--cc=namcao@linutronix.de \
--cc=npiggin@gmail.com \
--cc=oleg@redhat.com \
--cc=peterz@infradead.org \
--cc=segher@kernel.crashing.org \
--cc=sshegde@linux.ibm.com \
--cc=tglx@linutronix.de \
--cc=thomas.weissschuh@linutronix.de \
--cc=thuth@redhat.com \
--cc=wad@chromium.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).