All of lore.kernel.org
 help / color / mirror / Atom feed
From: Drew Fustini <fustini@kernel.org>
To: "Luck, Tony" <tony.luck@intel.com>
Cc: alex@ghiti.fr, aou@eecs.berkeley.edu, babu.moger@amd.com,
	baolin.wang@linux.alibaba.com, ben.horgan@arm.com,
	corbet@lwn.net, cuiyunhui@bytedance.com, Dave.Martin@arm.com,
	james.morse@arm.com, jgg@ziepe.ca, joro@8bytes.org,
	kevin.tian@intel.com, palmer@dabbelt.com, pjw@kernel.org,
	reinette.chatre@intel.com, robin.murphy@arm.com,
	shuah@kernel.org, tomasz.jeznach@linux.dev,
	wangzhou1@hisilicon.com, will@kernel.org, xiaqinxin@huawei.com,
	yuanzhu@bytedance.com, zengheng4@huawei.com,
	zhangzhanpeng.jasper@bytedance.com, x86@kernel.org,
	iommu@lists.linux.dev, linuxarm@huawei.com,
	linux-kernel@vger.kernel.org, linux-riscv@lists.infradead.org
Subject: Re: Resctrl and I/O
Date: Tue, 22 Sep 2026 18:27:51 -0700	[thread overview]
Message-ID: <arMrF51oJAXJrikO@thelio> (raw)
In-Reply-To: <arKx__kOPe4B5gDW@agluck-desk3>

On Tue, Sep 22, 2026 at 09:51:11AM -0700, Luck, Tony wrote:
> Patches for ARM[1] and RISC-V[2] are both at RFC status.
> 
> They appear to be based on similar h/w capabilities where architecture
> specific tags can be added to transactions from I/O devices by the IOMMU
> servicing the device. Device enumeration is handled outside of the resctrl
> file system.

Thanks for starting a discussion on this topic.

> RISC-V description of "architecture specific device tokens" sounds great.
> I'll have to find some ares of /sys to provide the mapping of tokens to
> "device groups". There's no association with IOMMUs, so this will be some
> other path in /sys. Perhaps under /sys/firmware/acpi since the data comes
> from ACPI IRDT table.

resctrl could parse iommu_group:<id> itself which should cover the IOMMU
case on ARM and RISC-V. Maybe other prefixes could be passed on to arch
code?

> Need an info file to tell users the format and source for device tokens.

I agree and it could list which token namespaces the platform uses.

> I'd like to avoid limiting resctrl monitoring of CPU activity due to the small
> number of I/O RMIDs. Perhaps don't include a "devices" file or mon_data/{i/o
> events} files in resctrl group directories when rdtgroup::mon::rmid >=
> num_io_rmids? But that feels clumsy.

There's no separate I/O ID space on RISC-V that could be smaller. The
IDs a device carries are the same IDs the hart (e.g. cpu) carries. The
number of RCIDs (e.g. CLOSID) and MCIDs (e.g. RMID) is a property of
each capacity or bandwidth (CBQRI) controller. The RISC-V IOMMU spec
also requires the IOMMU to support RCID and MCID at least as wide as any
hart.

Thanks,
Drew

WARNING: multiple messages have this Message-ID (diff)
From: Drew Fustini <fustini@kernel.org>
To: "Luck, Tony" <tony.luck@intel.com>
Cc: alex@ghiti.fr, aou@eecs.berkeley.edu, babu.moger@amd.com,
	baolin.wang@linux.alibaba.com, ben.horgan@arm.com,
	corbet@lwn.net, cuiyunhui@bytedance.com, Dave.Martin@arm.com,
	james.morse@arm.com, jgg@ziepe.ca, joro@8bytes.org,
	kevin.tian@intel.com, palmer@dabbelt.com, pjw@kernel.org,
	reinette.chatre@intel.com, robin.murphy@arm.com,
	shuah@kernel.org, tomasz.jeznach@linux.dev,
	wangzhou1@hisilicon.com, will@kernel.org, xiaqinxin@huawei.com,
	yuanzhu@bytedance.com, zengheng4@huawei.com,
	zhangzhanpeng.jasper@bytedance.com, x86@kernel.org,
	iommu@lists.linux.dev, linuxarm@huawei.com,
	linux-kernel@vger.kernel.org, linux-riscv@lists.infradead.org
Subject: Re: Resctrl and I/O
Date: Tue, 22 Sep 2026 18:27:51 -0700	[thread overview]
Message-ID: <arMrF51oJAXJrikO@thelio> (raw)
In-Reply-To: <arKx__kOPe4B5gDW@agluck-desk3>

On Tue, Sep 22, 2026 at 09:51:11AM -0700, Luck, Tony wrote:
> Patches for ARM[1] and RISC-V[2] are both at RFC status.
> 
> They appear to be based on similar h/w capabilities where architecture
> specific tags can be added to transactions from I/O devices by the IOMMU
> servicing the device. Device enumeration is handled outside of the resctrl
> file system.

Thanks for starting a discussion on this topic.

> RISC-V description of "architecture specific device tokens" sounds great.
> I'll have to find some ares of /sys to provide the mapping of tokens to
> "device groups". There's no association with IOMMUs, so this will be some
> other path in /sys. Perhaps under /sys/firmware/acpi since the data comes
> from ACPI IRDT table.

resctrl could parse iommu_group:<id> itself which should cover the IOMMU
case on ARM and RISC-V. Maybe other prefixes could be passed on to arch
code?

> Need an info file to tell users the format and source for device tokens.

I agree and it could list which token namespaces the platform uses.

> I'd like to avoid limiting resctrl monitoring of CPU activity due to the small
> number of I/O RMIDs. Perhaps don't include a "devices" file or mon_data/{i/o
> events} files in resctrl group directories when rdtgroup::mon::rmid >=
> num_io_rmids? But that feels clumsy.

There's no separate I/O ID space on RISC-V that could be smaller. The
IDs a device carries are the same IDs the hart (e.g. cpu) carries. The
number of RCIDs (e.g. CLOSID) and MCIDs (e.g. RMID) is a property of
each capacity or bandwidth (CBQRI) controller. The RISC-V IOMMU spec
also requires the IOMMU to support RCID and MCID at least as wide as any
hart.

Thanks,
Drew

_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

  reply	other threads:[~2026-09-23  1:27 UTC|newest]

Thread overview: 16+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-09-22 16:51 Resctrl and I/O Luck, Tony
2026-09-22 16:51 ` Luck, Tony
2026-09-23  1:27 ` Drew Fustini [this message]
2026-09-23  1:27   ` Drew Fustini
2026-09-23 17:13 ` James Morse
2026-09-23 17:13   ` James Morse
2026-09-23 17:24   ` Jason Gunthorpe
2026-09-23 17:24     ` Jason Gunthorpe
2026-09-24  2:30   ` [External] " yunhui cui
2026-09-24  2:30     ` yunhui cui
2026-09-24  9:38   ` Qinxin Xia
2026-09-24  9:38     ` Qinxin Xia
2026-09-24  9:40   ` Ben Horgan
2026-09-24  9:40     ` Ben Horgan
2026-09-24  9:51 ` Zhanpeng Zhang
2026-09-24  9:51   ` Zhanpeng Zhang

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=arMrF51oJAXJrikO@thelio \
    --to=fustini@kernel.org \
    --cc=Dave.Martin@arm.com \
    --cc=alex@ghiti.fr \
    --cc=aou@eecs.berkeley.edu \
    --cc=babu.moger@amd.com \
    --cc=baolin.wang@linux.alibaba.com \
    --cc=ben.horgan@arm.com \
    --cc=corbet@lwn.net \
    --cc=cuiyunhui@bytedance.com \
    --cc=iommu@lists.linux.dev \
    --cc=james.morse@arm.com \
    --cc=jgg@ziepe.ca \
    --cc=joro@8bytes.org \
    --cc=kevin.tian@intel.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-riscv@lists.infradead.org \
    --cc=linuxarm@huawei.com \
    --cc=palmer@dabbelt.com \
    --cc=pjw@kernel.org \
    --cc=reinette.chatre@intel.com \
    --cc=robin.murphy@arm.com \
    --cc=shuah@kernel.org \
    --cc=tomasz.jeznach@linux.dev \
    --cc=tony.luck@intel.com \
    --cc=wangzhou1@hisilicon.com \
    --cc=will@kernel.org \
    --cc=x86@kernel.org \
    --cc=xiaqinxin@huawei.com \
    --cc=yuanzhu@bytedance.com \
    --cc=zengheng4@huawei.com \
    --cc=zhangzhanpeng.jasper@bytedance.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.