public inbox for opensbi@lists.infradead.org
 help / color / mirror / Atom feed
From: Samuel Holland <samuel.holland@sifive.com>
To: dave.patel@riscstar.com, opensbi@lists.infradead.org
Cc: Scott Bambrough <scott@riscstar.com>,
	Robin Randhawa <robin.randhawa@sifive.com>,
	Anup Patel <anup.patel@qti.qualcomm.com>,
	Ray Mao <raymond.mao@riscstar.com>,
	Anup Patel <anuppate@qti.qualcomm.com>,
	Dhaval <dhaval@rivosinc.com>, Peter Lin <peter.lin@sifive.com>
Subject: Re: [PATCH 1/4] lib: sbi: Add RISC-V vector context save/restore support (eager switching)
Date: Sun, 22 Mar 2026 23:37:13 -0500	[thread overview]
Message-ID: <cfbfa618-d875-43e9-bbff-2aa8d75a519b@sifive.com> (raw)
In-Reply-To: <20260320143018.74191-2-dave.patel@riscstar.com>

Hi Dave,

On 2026-03-20 9:30 AM, dave.patel@riscstar.com wrote:
> From: Dave Patel <dave.patel@riscstar.com>
> 
> Add support for saving and restoring RISC-V vector extension state in OpenSBI.
> This introduces a per-hart vector context structure and helper routines to
> perform full context save and restore.
> 
> The vector context includes vl, vtype, vcsr CSRs along with storage for all
> 32 vector registers. The register state is saved and restored using byte-wise
> vector load/store instructions (vse8.v/vle8.v), making the implementation
> independent of current SEW/LMUL configuration.
> 
> The implementation follows an eager context switching model where the entire
> vector state is saved and restored on every context switch. This provides a
> simple and deterministic mechanism without requiring lazy trap-based
> management.
> 
> A per-hart pointer is used to track the current vector context owner.

Storing only one context per hart means that you must save/restore the context
every time the hart traps to M-mode. This is unnecessary and inefficient,
because OpenSBI itself does not use the vector registers (except where it
explicitly modifies the lower-privilege vector state when emulating
instructions). It is only necessary to save/restore the context when switching
between domains, not when trapping to M-mode and returning to the same domain.
So a better place for the context is in `struct hart_context` in
lib/sbi/sbi_domain_context.c.

> Notes:
> - The maximum supported VLEN is capped via SBI_MAX_VLENB.
> - The implementation assumes the vector unit is enabled when invoked.

This is not something you can assume. The code must not crash when running on
hardware without vector ISA support (Zve32x).

> - vstart CSR is not currently saved/restored and is expected to be zero
> across context switch boundaries.

This is not something you can assume. A trap can happen anywhere (e.g. due to an
interrupt), so vstart may be nonzero and must not leak across domains.

> 
> Signed-off-by: Dave Patel <dave.patel@riscstar.com>
> ---
>  include/sbi/sbi_vector.h |  30 +++++++++
>  lib/sbi/objects.mk       |   1 +
>  lib/sbi/sbi_vector.c     | 136 +++++++++++++++++++++++++++++++++++++++
>  3 files changed, 167 insertions(+)
>  create mode 100644 include/sbi/sbi_vector.h
>  create mode 100644 lib/sbi/sbi_vector.c
> 
> diff --git a/include/sbi/sbi_vector.h b/include/sbi/sbi_vector.h
> new file mode 100644
> index 00000000..6be477c0
> --- /dev/null
> +++ b/include/sbi/sbi_vector.h
> @@ -0,0 +1,30 @@
> +/* SPDX-License-Identifier: GPL-2.0
> + *
> + * Copyright (c) 2026 RISCstar Solutions.
> + *
> + * Authors:
> + *     Dave Patel <dave.patel@riscstar.com>
> + */
> +
> +#ifndef __SBI_VECTOR_H__
> +#define __SBI_VECTOR_H__
> +
> +#include <sbi/sbi_types.h>
> +
> +#define SBI_MAX_VLENB 256
> +
> +struct sbi_vector_context {
> +    unsigned long vl;
> +    unsigned long vtype;
> +    unsigned long vcsr;
> +
> +    /* size depends on VLEN */
> +    uint8_t vregs[32 * SBI_MAX_VLENB];
> +};
> +
> +struct sbi_vector_context *sbi_current_vector_context(void);
> +void sbi_vector_save(struct sbi_vector_context *dst);
> +void sbi_vector_restore(const struct sbi_vector_context *src);
> +
> +#endif //__SBI_VECTOR_H__
> +
> diff --git a/lib/sbi/objects.mk b/lib/sbi/objects.mk
> index ea816e92..5c0caf39 100644
> --- a/lib/sbi/objects.mk
> +++ b/lib/sbi/objects.mk
> @@ -106,3 +106,4 @@ libsbi-objs-y += sbi_trap_v_ldst.o
>  libsbi-objs-y += sbi_unpriv.o
>  libsbi-objs-y += sbi_expected_trap.o
>  libsbi-objs-y += sbi_cppc.o
> +libsbi-objs-y += sbi_vector.o
> diff --git a/lib/sbi/sbi_vector.c b/lib/sbi/sbi_vector.c
> new file mode 100644
> index 00000000..37239ce2
> --- /dev/null
> +++ b/lib/sbi/sbi_vector.c
> @@ -0,0 +1,136 @@
> +/* SPDX-License-Identifier: GPL-2.0
> + *
> + * Copyright (c) 2026 RISCstar Solutions.
> + *
> + * Authors:
> + *     Dave Patel <dave.patel@riscstar.com>
> + */
> +
> +#include <sbi/sbi_domain.h>
> +#include <sbi/riscv_encoding.h>
> +#include <sbi/riscv_asm.h>
> +#include <sbi/sbi_vector.h>
> +
> +/* Per-hart vector owner */
> +static inline struct sbi_vector_context **vec_owner_ptr(void)
> +{
> +    struct sbi_scratch *scratch = sbi_scratch_thishart_ptr();
> +    return &scratch->vec_ctx;

The code must compile and run after every patch in the series, but the vec_ctx
member isn't added until patch 3, so this doesn't compile. However, as mentioned
above, the context should be stored per-domain anyway.

Regards,
Samuel

> +}
> +
> +/* Get current vector context */
> +struct sbi_vector_context *sbi_current_vector_context(void)
> +{
> +    return *vec_owner_ptr();
> +}
> +
> +static inline unsigned long vector_vlenb(void)
> +{
> +    unsigned long vlenb;
> +    asm volatile ("csrr %0, vlenb" : "=r"(vlenb));
> +    return vlenb;
> +}
> +
> +
> +void sbi_vector_save(struct sbi_vector_context *dst)
> +{
> +    if (!dst)
> +        return;
> +
> +    uint8_t *base = dst->vregs;
> +    unsigned long vlenb = vector_vlenb();
> +
> +    asm volatile("csrr %0, vtype" : "=r"(dst->vtype));
> +    asm volatile("csrr %0, vl" : "=r"(dst->vl));
> +    asm volatile("csrr %0, vcsr" : "=r"(dst->vcsr));
> +
> +#define SAVE_VREG(num) \
> +    asm volatile("vse8.v v" #num ", (%0)" :: "r"(base + num * vlenb) : "memory")
> +
> +    SAVE_VREG(0);
> +    SAVE_VREG(1);
> +    SAVE_VREG(2);
> +    SAVE_VREG(3);
> +    SAVE_VREG(4);
> +    SAVE_VREG(5);
> +    SAVE_VREG(6);
> +    SAVE_VREG(7);
> +    SAVE_VREG(8);
> +    SAVE_VREG(9);
> +    SAVE_VREG(10);
> +    SAVE_VREG(11);
> +    SAVE_VREG(12);
> +    SAVE_VREG(13);
> +    SAVE_VREG(14);
> +    SAVE_VREG(15);
> +    SAVE_VREG(16);
> +    SAVE_VREG(17);
> +    SAVE_VREG(18);
> +    SAVE_VREG(19);
> +    SAVE_VREG(20);
> +    SAVE_VREG(21);
> +    SAVE_VREG(22);
> +    SAVE_VREG(23);
> +    SAVE_VREG(24);
> +    SAVE_VREG(25);
> +    SAVE_VREG(26);
> +    SAVE_VREG(27);
> +    SAVE_VREG(28);
> +    SAVE_VREG(29);
> +    SAVE_VREG(30);
> +    SAVE_VREG(31);
> +
> +#undef SAVE_VREG
> +}
> +
> +void sbi_vector_restore(const struct sbi_vector_context *src)
> +{
> +    if (!src)
> +        return;
> +
> +    const uint8_t *base = src->vregs;
> +    unsigned long vlenb = vector_vlenb();
> +
> +    asm volatile("csrw vtype, %0" :: "r"(src->vtype));
> +    asm volatile("csrw vl, %0" :: "r"(src->vl));
> +    asm volatile("csrw vcsr, %0" :: "r"(src->vcsr));
> +
> +#define RESTORE_VREG(num) \
> +    asm volatile("vle8.v v" #num ", (%0)" :: "r"(base + num * vlenb) : "memory")
> +
> +    RESTORE_VREG(0);
> +    RESTORE_VREG(1);
> +    RESTORE_VREG(2);
> +    RESTORE_VREG(3);
> +    RESTORE_VREG(4);
> +    RESTORE_VREG(5);
> +    RESTORE_VREG(6);
> +    RESTORE_VREG(7);
> +    RESTORE_VREG(8);
> +    RESTORE_VREG(9);
> +    RESTORE_VREG(10);
> +    RESTORE_VREG(11);
> +    RESTORE_VREG(12);
> +    RESTORE_VREG(13);
> +    RESTORE_VREG(14);
> +    RESTORE_VREG(15);
> +    RESTORE_VREG(16);
> +    RESTORE_VREG(17);
> +    RESTORE_VREG(18);
> +    RESTORE_VREG(19);
> +    RESTORE_VREG(20);
> +    RESTORE_VREG(21);
> +    RESTORE_VREG(22);
> +    RESTORE_VREG(23);
> +    RESTORE_VREG(24);
> +    RESTORE_VREG(25);
> +    RESTORE_VREG(26);
> +    RESTORE_VREG(27);
> +    RESTORE_VREG(28);
> +    RESTORE_VREG(29);
> +    RESTORE_VREG(30);
> +    RESTORE_VREG(31);
> +
> +#undef RESTORE_VREG
> +}
> +


-- 
opensbi mailing list
opensbi@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/opensbi

  reply	other threads:[~2026-03-23  4:37 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-03-20 14:30 [PATCH v1 0/4] Add eager FP and RISC-V vector context switching support dave.patel
2026-03-20 14:30 ` [PATCH 1/4] lib: sbi: Add RISC-V vector context save/restore support (eager switching) dave.patel
2026-03-23  4:37   ` Samuel Holland [this message]
2026-03-23 10:05     ` Dave Patel
2026-03-20 14:30 ` [PATCH 2/4] lib: sbi: Add floating-point context save/restore support dave.patel
2026-03-20 14:30 ` [PATCH 3/4] include: sbi: scratch: Add per-hart FP and vector context pointers in scratch dave.patel
2026-03-20 14:30 ` [PATCH 4/4] lib: sbi: trap: Save/restore FP and vector state on trap entry/exit dave.patel
2026-03-23  4:42   ` Samuel Holland

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=cfbfa618-d875-43e9-bbff-2aa8d75a519b@sifive.com \
    --to=samuel.holland@sifive.com \
    --cc=anup.patel@qti.qualcomm.com \
    --cc=anuppate@qti.qualcomm.com \
    --cc=dave.patel@riscstar.com \
    --cc=dhaval@rivosinc.com \
    --cc=opensbi@lists.infradead.org \
    --cc=peter.lin@sifive.com \
    --cc=raymond.mao@riscstar.com \
    --cc=robin.randhawa@sifive.com \
    --cc=scott@riscstar.com \
    /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