From: Charlie Jenkins <charlie@rivosinc.com>
To: "Clément Léger" <cleger@rivosinc.com>
Cc: Paul Walmsley <paul.walmsley@sifive.com>,
Palmer Dabbelt <palmer@dabbelt.com>,
Albert Ou <aou@eecs.berkeley.edu>,
Andrea Parri <parri.andrea@gmail.com>,
linux-kernel@vger.kernel.org, linux-riscv@lists.infradead.org
Subject: Re: [PATCH 1/2] tools: Add riscv barrier implementation
Date: Wed, 31 Jul 2024 18:10:55 -0700 [thread overview]
Message-ID: <Zqrgn04zEmEpv5+m@ghost> (raw)
In-Reply-To: <28d780da-4f43-4db5-8e1b-c66bb9973cd1@rivosinc.com>
On Tue, Jul 30, 2024 at 10:59:52AM +0200, Clément Léger wrote:
>
>
> On 29/07/2024 22:50, Charlie Jenkins wrote:
> > Many of the other architectures use their custom barrier implmentations.
>
> Hi Charlie,
>
> Typo: implmentations -> implementations
Thank you! I will fix that.
- Charlie
>
> > Use the barrier code from the kernel sources to optimize barriers in
> > tools.
> >
> > Signed-off-by: Charlie Jenkins <charlie@rivosinc.com>
> > ---
> > tools/arch/riscv/include/asm/barrier.h | 39 ++++++++++++++++++++++++++++++++++
> > tools/arch/riscv/include/asm/fence.h | 13 ++++++++++++
> > tools/include/asm/barrier.h | 2 ++
> > 3 files changed, 54 insertions(+)
> >
> > diff --git a/tools/arch/riscv/include/asm/barrier.h b/tools/arch/riscv/include/asm/barrier.h
> > new file mode 100644
> > index 000000000000..6997f197086d
> > --- /dev/null
> > +++ b/tools/arch/riscv/include/asm/barrier.h
> > @@ -0,0 +1,39 @@
> > +/* SPDX-License-Identifier: GPL-2.0-only */
> > +/*
> > + * Copied from the kernel sources to tools/arch/riscv:
> > + *
> > + * Copyright (C) 2012 ARM Ltd.
> > + * Copyright (C) 2013 Regents of the University of California
> > + * Copyright (C) 2017 SiFive
> > + */
> > +
> > +#ifndef _TOOLS_LINUX_ASM_RISCV_BARRIER_H
> > +#define _TOOLS_LINUX_ASM_RISCV_BARRIER_H
> > +
> > +#include <asm/fence.h>
> > +#include <linux/compiler.h>
> > +
> > +/* These barriers need to enforce ordering on both devices and memory. */
> > +#define mb() RISCV_FENCE(iorw, iorw)
> > +#define rmb() RISCV_FENCE(ir, ir)
> > +#define wmb() RISCV_FENCE(ow, ow)
> > +
> > +/* These barriers do not need to enforce ordering on devices, just memory. */
> > +#define smp_mb() RISCV_FENCE(rw, rw)
> > +#define smp_rmb() RISCV_FENCE(r, r)
> > +#define smp_wmb() RISCV_FENCE(w, w)
> > +
> > +#define smp_store_release(p, v) \
> > +do { \
> > + RISCV_FENCE(rw, w); \
> > + WRITE_ONCE(*p, v); \
> > +} while (0)
> > +
> > +#define smp_load_acquire(p) \
> > +({ \
> > + typeof(*p) ___p1 = READ_ONCE(*p); \
> > + RISCV_FENCE(r, rw); \
> > + ___p1; \
> > +})
> > +
> > +#endif /* _TOOLS_LINUX_ASM_RISCV_BARRIER_H */
> > diff --git a/tools/arch/riscv/include/asm/fence.h b/tools/arch/riscv/include/asm/fence.h
> > new file mode 100644
> > index 000000000000..37860e86771d
> > --- /dev/null
> > +++ b/tools/arch/riscv/include/asm/fence.h
> > @@ -0,0 +1,13 @@
> > +/* SPDX-License-Identifier: GPL-2.0-only */
> > +/*
> > + * Copied from the kernel sources to tools/arch/riscv:
> > + */
> > +
> > +#ifndef _ASM_RISCV_FENCE_H
> > +#define _ASM_RISCV_FENCE_H
> > +
> > +#define RISCV_FENCE_ASM(p, s) "\tfence " #p "," #s "\n"
> > +#define RISCV_FENCE(p, s) \
> > + ({ __asm__ __volatile__ (RISCV_FENCE_ASM(p, s) : : : "memory"); })
> > +
> > +#endif /* _ASM_RISCV_FENCE_H */
> > diff --git a/tools/include/asm/barrier.h b/tools/include/asm/barrier.h
> > index 8d378c57cb01..0c21678ac5e6 100644
> > --- a/tools/include/asm/barrier.h
> > +++ b/tools/include/asm/barrier.h
> > @@ -8,6 +8,8 @@
> > #include "../../arch/arm64/include/asm/barrier.h"
> > #elif defined(__powerpc__)
> > #include "../../arch/powerpc/include/asm/barrier.h"
> > +#elif defined(__riscv)
> > +#include "../../arch/riscv/include/asm/barrier.h"
> > #elif defined(__s390__)
> > #include "../../arch/s390/include/asm/barrier.h"
> > #elif defined(__sh__)
> >
>
> Can not really tell for that part except it seems ok to me as well.
> Andrea might be a better candidate to add its Rb.
>
> Thanks,
>
> Clément
_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv
WARNING: multiple messages have this Message-ID (diff)
From: Charlie Jenkins <charlie@rivosinc.com>
To: "Clément Léger" <cleger@rivosinc.com>
Cc: Paul Walmsley <paul.walmsley@sifive.com>,
Palmer Dabbelt <palmer@dabbelt.com>,
Albert Ou <aou@eecs.berkeley.edu>,
Andrea Parri <parri.andrea@gmail.com>,
linux-kernel@vger.kernel.org, linux-riscv@lists.infradead.org
Subject: Re: [PATCH 1/2] tools: Add riscv barrier implementation
Date: Wed, 31 Jul 2024 18:10:55 -0700 [thread overview]
Message-ID: <Zqrgn04zEmEpv5+m@ghost> (raw)
In-Reply-To: <28d780da-4f43-4db5-8e1b-c66bb9973cd1@rivosinc.com>
On Tue, Jul 30, 2024 at 10:59:52AM +0200, Clément Léger wrote:
>
>
> On 29/07/2024 22:50, Charlie Jenkins wrote:
> > Many of the other architectures use their custom barrier implmentations.
>
> Hi Charlie,
>
> Typo: implmentations -> implementations
Thank you! I will fix that.
- Charlie
>
> > Use the barrier code from the kernel sources to optimize barriers in
> > tools.
> >
> > Signed-off-by: Charlie Jenkins <charlie@rivosinc.com>
> > ---
> > tools/arch/riscv/include/asm/barrier.h | 39 ++++++++++++++++++++++++++++++++++
> > tools/arch/riscv/include/asm/fence.h | 13 ++++++++++++
> > tools/include/asm/barrier.h | 2 ++
> > 3 files changed, 54 insertions(+)
> >
> > diff --git a/tools/arch/riscv/include/asm/barrier.h b/tools/arch/riscv/include/asm/barrier.h
> > new file mode 100644
> > index 000000000000..6997f197086d
> > --- /dev/null
> > +++ b/tools/arch/riscv/include/asm/barrier.h
> > @@ -0,0 +1,39 @@
> > +/* SPDX-License-Identifier: GPL-2.0-only */
> > +/*
> > + * Copied from the kernel sources to tools/arch/riscv:
> > + *
> > + * Copyright (C) 2012 ARM Ltd.
> > + * Copyright (C) 2013 Regents of the University of California
> > + * Copyright (C) 2017 SiFive
> > + */
> > +
> > +#ifndef _TOOLS_LINUX_ASM_RISCV_BARRIER_H
> > +#define _TOOLS_LINUX_ASM_RISCV_BARRIER_H
> > +
> > +#include <asm/fence.h>
> > +#include <linux/compiler.h>
> > +
> > +/* These barriers need to enforce ordering on both devices and memory. */
> > +#define mb() RISCV_FENCE(iorw, iorw)
> > +#define rmb() RISCV_FENCE(ir, ir)
> > +#define wmb() RISCV_FENCE(ow, ow)
> > +
> > +/* These barriers do not need to enforce ordering on devices, just memory. */
> > +#define smp_mb() RISCV_FENCE(rw, rw)
> > +#define smp_rmb() RISCV_FENCE(r, r)
> > +#define smp_wmb() RISCV_FENCE(w, w)
> > +
> > +#define smp_store_release(p, v) \
> > +do { \
> > + RISCV_FENCE(rw, w); \
> > + WRITE_ONCE(*p, v); \
> > +} while (0)
> > +
> > +#define smp_load_acquire(p) \
> > +({ \
> > + typeof(*p) ___p1 = READ_ONCE(*p); \
> > + RISCV_FENCE(r, rw); \
> > + ___p1; \
> > +})
> > +
> > +#endif /* _TOOLS_LINUX_ASM_RISCV_BARRIER_H */
> > diff --git a/tools/arch/riscv/include/asm/fence.h b/tools/arch/riscv/include/asm/fence.h
> > new file mode 100644
> > index 000000000000..37860e86771d
> > --- /dev/null
> > +++ b/tools/arch/riscv/include/asm/fence.h
> > @@ -0,0 +1,13 @@
> > +/* SPDX-License-Identifier: GPL-2.0-only */
> > +/*
> > + * Copied from the kernel sources to tools/arch/riscv:
> > + */
> > +
> > +#ifndef _ASM_RISCV_FENCE_H
> > +#define _ASM_RISCV_FENCE_H
> > +
> > +#define RISCV_FENCE_ASM(p, s) "\tfence " #p "," #s "\n"
> > +#define RISCV_FENCE(p, s) \
> > + ({ __asm__ __volatile__ (RISCV_FENCE_ASM(p, s) : : : "memory"); })
> > +
> > +#endif /* _ASM_RISCV_FENCE_H */
> > diff --git a/tools/include/asm/barrier.h b/tools/include/asm/barrier.h
> > index 8d378c57cb01..0c21678ac5e6 100644
> > --- a/tools/include/asm/barrier.h
> > +++ b/tools/include/asm/barrier.h
> > @@ -8,6 +8,8 @@
> > #include "../../arch/arm64/include/asm/barrier.h"
> > #elif defined(__powerpc__)
> > #include "../../arch/powerpc/include/asm/barrier.h"
> > +#elif defined(__riscv)
> > +#include "../../arch/riscv/include/asm/barrier.h"
> > #elif defined(__s390__)
> > #include "../../arch/s390/include/asm/barrier.h"
> > #elif defined(__sh__)
> >
>
> Can not really tell for that part except it seems ok to me as well.
> Andrea might be a better candidate to add its Rb.
>
> Thanks,
>
> Clément
next prev parent reply other threads:[~2024-08-01 1:11 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-07-29 20:50 [PATCH 0/2] tools: Add barrier implementations for riscv Charlie Jenkins
2024-07-29 20:50 ` Charlie Jenkins
2024-07-29 20:50 ` [PATCH 1/2] tools: Add riscv barrier implementation Charlie Jenkins
2024-07-29 20:50 ` Charlie Jenkins
2024-07-30 8:59 ` Clément Léger
2024-07-30 8:59 ` Clément Léger
2024-08-01 1:10 ` Charlie Jenkins [this message]
2024-08-01 1:10 ` Charlie Jenkins
2024-07-29 20:50 ` [PATCH 2/2] tools: Optimize ring buffer for riscv Charlie Jenkins
2024-07-29 20:50 ` Charlie Jenkins
2024-07-30 9:18 ` [PATCH 0/2] tools: Add barrier implementations " Andrea Parri
2024-07-30 9:18 ` Andrea Parri
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=Zqrgn04zEmEpv5+m@ghost \
--to=charlie@rivosinc.com \
--cc=aou@eecs.berkeley.edu \
--cc=cleger@rivosinc.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-riscv@lists.infradead.org \
--cc=palmer@dabbelt.com \
--cc=parri.andrea@gmail.com \
--cc=paul.walmsley@sifive.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 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.