public inbox for iommu@lists.linux-foundation.org
 help / color / mirror / Atom feed
From: Alex Williamson <alex.williamson-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
To: Jacob Pan <jacob.jun.pan-VuQAYsv1563Yd54FQh9/CA@public.gmane.org>
Cc: "Tian,
	Kevin" <kevin.tian-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>,
	Raj Ashok <ashok.raj-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>,
	Jean-Philippe Brucker
	<jean-philippe.brucker-5wv7dgnIgG8@public.gmane.org>,
	LKML <linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>,
	iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org,
	Andriy Shevchenko
	<andriy.shevchenko-VuQAYsv1563Yd54FQh9/CA@public.gmane.org>,
	David Woodhouse <dwmw2-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>
Subject: Re: [PATCH 10/18] iommu/vt-d: Add custom allocator for IOASID
Date: Mon, 15 Apr 2019 14:37:11 -0600	[thread overview]
Message-ID: <20190415143711.13c29308@x1.home> (raw)
In-Reply-To: <1554767973-30125-11-git-send-email-jacob.jun.pan-VuQAYsv1563Yd54FQh9/CA@public.gmane.org>

On Mon,  8 Apr 2019 16:59:25 -0700
Jacob Pan <jacob.jun.pan-VuQAYsv1563Yd54FQh9/CA@public.gmane.org> wrote:

> When VT-d driver runs in the guest, PASID allocation must be
> performed via virtual command interface. This patch register a
> custom IOASID allocator which takes precedence over the default
> IDR based allocator. The resulting IOASID allocation will always
> come from the host. This ensures that PASID namespace is system-
> wide.
> 
> Signed-off-by: Lu Baolu <baolu.lu-VuQAYsv1563Yd54FQh9/CA@public.gmane.org>
> Signed-off-by: Liu, Yi L <yi.l.liu-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
> Signed-off-by: Jacob Pan <jacob.jun.pan-VuQAYsv1563Yd54FQh9/CA@public.gmane.org>
> ---
>  drivers/iommu/intel-iommu.c | 50 +++++++++++++++++++++++++++++++++++++++++++++
>  include/linux/intel-iommu.h |  1 +
>  2 files changed, 51 insertions(+)
> 
> diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c
> index 28cb713..a38d774 100644
> --- a/drivers/iommu/intel-iommu.c
> +++ b/drivers/iommu/intel-iommu.c
> @@ -4820,6 +4820,42 @@ static int __init platform_optin_force_iommu(void)
>  	return 1;
>  }
>  
> +static ioasid_t intel_ioasid_alloc(ioasid_t min, ioasid_t max, void *data)
> +{
> +	struct intel_iommu *iommu = data;
> +	ioasid_t ioasid;
> +
> +	if (vcmd_alloc_pasid(iommu, &ioasid))
> +		return INVALID_IOASID;
> +	return ioasid;

How does this honor min/max?

> +}
> +
> +static int intel_ioasid_free(ioasid_t ioasid, void *data)
> +{
> +	struct iommu_pasid_alloc_info *svm;
> +	struct intel_iommu *iommu = data;
> +
> +	if (!iommu || !cap_caching_mode(iommu->cap))
> +		return -EINVAL;
> +	/*
> +	 * Sanity check the ioasid owner is done at upper layer, e.g. VFIO
> +	 * We can only free the PASID when all the devices are unbond.
> +	 */
> +	svm = ioasid_find(NULL, ioasid, NULL);
> +	if (!svm) {
> +		pr_warn("Freeing unbond IOASID %d\n", ioasid);
> +		return -EBUSY;
> +	}
> +	vcmd_free_pasid(iommu, ioasid);
> +
> +	return 0;
> +}
> +
> +static struct ioasid_allocator intel_iommu_ioasid_allocator = {
> +	.alloc = intel_ioasid_alloc,
> +	.free = intel_ioasid_free,
> +};
> +
>  int __init intel_iommu_init(void)
>  {
>  	int ret = -ENODEV;
> @@ -4921,6 +4957,20 @@ int __init intel_iommu_init(void)
>  				       "%s", iommu->name);
>  		iommu_device_set_ops(&iommu->iommu, &intel_iommu_ops);
>  		iommu_device_register(&iommu->iommu);
> +		if (cap_caching_mode(iommu->cap) && sm_supported(iommu)) {
> +			/*
> +			 * Register a custom ASID allocator if we are running
> +			 * in a guest, the purpose is to have a system wide PASID
> +			 * namespace among all PASID users.
> +			 * Note that only one vIOMMU in each guest is supported.

Why one vIOMMU per guest?  This would prevent guests with multiple PCI
domains aiui.

> +			 */
> +			intel_iommu_ioasid_allocator.pdata = (void *)iommu;
> +			ret = ioasid_set_allocator(&intel_iommu_ioasid_allocator);
> +			if (ret == -EBUSY) {
> +				pr_info("Custom IOASID allocator already registered\n");
> +				break;
> +			}
> +		}
>  	}
>  
>  	bus_set_iommu(&pci_bus_type, &intel_iommu_ops);
> diff --git a/include/linux/intel-iommu.h b/include/linux/intel-iommu.h
> index b29c85c..bc09d80 100644
> --- a/include/linux/intel-iommu.h
> +++ b/include/linux/intel-iommu.h
> @@ -31,6 +31,7 @@
>  #include <linux/iommu.h>
>  #include <linux/io-64-nonatomic-lo-hi.h>
>  #include <linux/dmar.h>
> +#include <linux/ioasid.h>
>  
>  #include <asm/cacheflush.h>
>  #include <asm/iommu.h>

WARNING: multiple messages have this Message-ID (diff)
From: Alex Williamson <alex.williamson@redhat.com>
To: Jacob Pan <jacob.jun.pan@linux.intel.com>
Cc: "Tian, Kevin" <kevin.tian@intel.com>,
	Raj Ashok <ashok.raj@intel.com>,
	Jean-Philippe Brucker <jean-philippe.brucker@arm.com>,
	LKML <linux-kernel@vger.kernel.org>,
	iommu@lists.linux-foundation.org,
	Andriy Shevchenko <andriy.shevchenko@linux.intel.com>,
	David Woodhouse <dwmw2@infradead.org>
Subject: Re: [PATCH 10/18] iommu/vt-d: Add custom allocator for IOASID
Date: Mon, 15 Apr 2019 14:37:11 -0600	[thread overview]
Message-ID: <20190415143711.13c29308@x1.home> (raw)
Message-ID: <20190415203711.OOerlM1tACHPg20lHQxwzPmg7CdAxIxndFJD97yC5_U@z> (raw)
In-Reply-To: <1554767973-30125-11-git-send-email-jacob.jun.pan@linux.intel.com>

On Mon,  8 Apr 2019 16:59:25 -0700
Jacob Pan <jacob.jun.pan@linux.intel.com> wrote:

> When VT-d driver runs in the guest, PASID allocation must be
> performed via virtual command interface. This patch register a
> custom IOASID allocator which takes precedence over the default
> IDR based allocator. The resulting IOASID allocation will always
> come from the host. This ensures that PASID namespace is system-
> wide.
> 
> Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
> Signed-off-by: Liu, Yi L <yi.l.liu@intel.com>
> Signed-off-by: Jacob Pan <jacob.jun.pan@linux.intel.com>
> ---
>  drivers/iommu/intel-iommu.c | 50 +++++++++++++++++++++++++++++++++++++++++++++
>  include/linux/intel-iommu.h |  1 +
>  2 files changed, 51 insertions(+)
> 
> diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c
> index 28cb713..a38d774 100644
> --- a/drivers/iommu/intel-iommu.c
> +++ b/drivers/iommu/intel-iommu.c
> @@ -4820,6 +4820,42 @@ static int __init platform_optin_force_iommu(void)
>  	return 1;
>  }
>  
> +static ioasid_t intel_ioasid_alloc(ioasid_t min, ioasid_t max, void *data)
> +{
> +	struct intel_iommu *iommu = data;
> +	ioasid_t ioasid;
> +
> +	if (vcmd_alloc_pasid(iommu, &ioasid))
> +		return INVALID_IOASID;
> +	return ioasid;

How does this honor min/max?

> +}
> +
> +static int intel_ioasid_free(ioasid_t ioasid, void *data)
> +{
> +	struct iommu_pasid_alloc_info *svm;
> +	struct intel_iommu *iommu = data;
> +
> +	if (!iommu || !cap_caching_mode(iommu->cap))
> +		return -EINVAL;
> +	/*
> +	 * Sanity check the ioasid owner is done at upper layer, e.g. VFIO
> +	 * We can only free the PASID when all the devices are unbond.
> +	 */
> +	svm = ioasid_find(NULL, ioasid, NULL);
> +	if (!svm) {
> +		pr_warn("Freeing unbond IOASID %d\n", ioasid);
> +		return -EBUSY;
> +	}
> +	vcmd_free_pasid(iommu, ioasid);
> +
> +	return 0;
> +}
> +
> +static struct ioasid_allocator intel_iommu_ioasid_allocator = {
> +	.alloc = intel_ioasid_alloc,
> +	.free = intel_ioasid_free,
> +};
> +
>  int __init intel_iommu_init(void)
>  {
>  	int ret = -ENODEV;
> @@ -4921,6 +4957,20 @@ int __init intel_iommu_init(void)
>  				       "%s", iommu->name);
>  		iommu_device_set_ops(&iommu->iommu, &intel_iommu_ops);
>  		iommu_device_register(&iommu->iommu);
> +		if (cap_caching_mode(iommu->cap) && sm_supported(iommu)) {
> +			/*
> +			 * Register a custom ASID allocator if we are running
> +			 * in a guest, the purpose is to have a system wide PASID
> +			 * namespace among all PASID users.
> +			 * Note that only one vIOMMU in each guest is supported.

Why one vIOMMU per guest?  This would prevent guests with multiple PCI
domains aiui.

> +			 */
> +			intel_iommu_ioasid_allocator.pdata = (void *)iommu;
> +			ret = ioasid_set_allocator(&intel_iommu_ioasid_allocator);
> +			if (ret == -EBUSY) {
> +				pr_info("Custom IOASID allocator already registered\n");
> +				break;
> +			}
> +		}
>  	}
>  
>  	bus_set_iommu(&pci_bus_type, &intel_iommu_ops);
> diff --git a/include/linux/intel-iommu.h b/include/linux/intel-iommu.h
> index b29c85c..bc09d80 100644
> --- a/include/linux/intel-iommu.h
> +++ b/include/linux/intel-iommu.h
> @@ -31,6 +31,7 @@
>  #include <linux/iommu.h>
>  #include <linux/io-64-nonatomic-lo-hi.h>
>  #include <linux/dmar.h>
> +#include <linux/ioasid.h>
>  
>  #include <asm/cacheflush.h>
>  #include <asm/iommu.h>

_______________________________________________
iommu mailing list
iommu@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/iommu

  parent reply	other threads:[~2019-04-15 20:37 UTC|newest]

Thread overview: 98+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-04-08 23:59 [PATCH 00/18] Shared virtual address IOMMU and VT-d support Jacob Pan
2019-04-08 23:59 ` Jacob Pan
2019-04-08 23:59 ` [PATCH 01/18] drivers core: Add I/O ASID allocator Jacob Pan
2019-04-08 23:59   ` Jacob Pan
     [not found]   ` <1554767973-30125-2-git-send-email-jacob.jun.pan-VuQAYsv1563Yd54FQh9/CA@public.gmane.org>
2019-04-09 10:00     ` Andriy Shevchenko
2019-04-09 10:00       ` Andriy Shevchenko
     [not found]       ` <20190409100049.GC9224-XvqNBM/wLWRrdx17CPfAsdBPR1lH4CV8@public.gmane.org>
2019-04-09 10:04         ` Christoph Hellwig
2019-04-09 10:04           ` Christoph Hellwig
2019-04-09 10:30           ` Andriy Shevchenko
2019-04-09 10:30             ` Andriy Shevchenko
2019-04-09 14:53             ` Paul E. McKenney
2019-04-09 14:53               ` Paul E. McKenney
2019-04-09 15:21               ` Andriy Shevchenko
2019-04-09 15:21                 ` Andriy Shevchenko
2019-04-09 22:08                 ` Paul E. McKenney
2019-04-09 22:08                   ` Paul E. McKenney
2019-04-08 23:59 ` [PATCH 05/18] iommu: introduce device fault data Jacob Pan
2019-04-08 23:59   ` Jacob Pan
2019-04-09 10:03   ` Andriy Shevchenko
2019-04-09 10:03     ` Andriy Shevchenko
     [not found]     ` <20190409100315.GD9224-XvqNBM/wLWRrdx17CPfAsdBPR1lH4CV8@public.gmane.org>
2019-04-09 16:44       ` Jacob Pan
2019-04-09 16:44         ` Jacob Pan
2019-04-08 23:59 ` [PATCH 06/18] iommu: introduce device fault report API Jacob Pan
2019-04-08 23:59   ` Jacob Pan
2019-04-08 23:59 ` [PATCH 10/18] iommu/vt-d: Add custom allocator for IOASID Jacob Pan
2019-04-08 23:59   ` Jacob Pan
     [not found]   ` <1554767973-30125-11-git-send-email-jacob.jun.pan-VuQAYsv1563Yd54FQh9/CA@public.gmane.org>
2019-04-15 20:37     ` Alex Williamson [this message]
2019-04-15 20:37       ` Alex Williamson
2019-04-15 23:10       ` Jacob Pan
2019-04-15 23:10         ` Jacob Pan
2019-04-18 15:36         ` Jean-Philippe Brucker
2019-04-18 15:36           ` Jean-Philippe Brucker
     [not found]           ` <70c6b197-029e-94cd-2745-d4a193cbbcd0-5wv7dgnIgG8@public.gmane.org>
2019-04-19  4:29             ` Jacob Pan
2019-04-19  4:29               ` Jacob Pan
2019-04-23 10:53               ` Jean-Philippe Brucker
2019-04-23 10:53                 ` Jean-Philippe Brucker
2019-04-16 15:30       ` Jacob Pan
2019-04-16 15:30         ` Jacob Pan
2019-04-08 23:59 ` [PATCH 11/18] iommu/vt-d: Replace Intel specific PASID allocator with IOASID Jacob Pan
2019-04-08 23:59   ` Jacob Pan
2019-04-08 23:59 ` [PATCH 12/18] iommu: Add guest PASID bind function Jacob Pan
2019-04-08 23:59   ` Jacob Pan
2019-04-08 23:59 ` [PATCH 13/18] iommu/vt-d: Move domain helper to header Jacob Pan
2019-04-08 23:59   ` Jacob Pan
2019-04-08 23:59 ` [PATCH 14/18] iommu/vt-d: Add nested translation support Jacob Pan
2019-04-08 23:59   ` Jacob Pan
2019-04-08 23:59 ` [PATCH 15/18] iommu/vt-d: Add bind guest PASID support Jacob Pan
2019-04-08 23:59   ` Jacob Pan
     [not found]   ` <1554767973-30125-16-git-send-email-jacob.jun.pan-VuQAYsv1563Yd54FQh9/CA@public.gmane.org>
2019-04-09 14:52     ` Andriy Shevchenko
2019-04-09 14:52       ` Andriy Shevchenko
2019-04-08 23:59 ` [PATCH 16/18] iommu: add max num of cache and granu types Jacob Pan
2019-04-08 23:59   ` Jacob Pan
2019-04-09 14:53   ` Andriy Shevchenko
2019-04-09 14:53     ` Andriy Shevchenko
2019-04-08 23:59 ` [PATCH 17/18] iommu/vt-d: Support flushing more translation cache types Jacob Pan
2019-04-08 23:59   ` Jacob Pan
2019-04-08 23:59 ` [PATCH 18/18] iommu/vt-d: Add svm/sva invalidate function Jacob Pan
2019-04-08 23:59   ` Jacob Pan
2019-04-09 14:57   ` Andriy Shevchenko
2019-04-09 14:57     ` Andriy Shevchenko
2019-04-09 17:43     ` Jacob Pan
2019-04-09 17:43       ` Jacob Pan
2019-04-09  9:56 ` [PATCH 00/18] Shared virtual address IOMMU and VT-d support Andriy Shevchenko
2019-04-09  9:56   ` Andriy Shevchenko
     [not found]   ` <20190409095623.GB9224-XvqNBM/wLWRrdx17CPfAsdBPR1lH4CV8@public.gmane.org>
2019-04-09 16:33     ` Jacob Pan
2019-04-09 16:33       ` Jacob Pan
     [not found] ` <1554767973-30125-1-git-send-email-jacob.jun.pan-VuQAYsv1563Yd54FQh9/CA@public.gmane.org>
2019-04-08 23:59   ` [PATCH 02/18] ioasid: Add custom IOASID allocator Jacob Pan
2019-04-08 23:59     ` Jacob Pan
2019-04-15 18:53     ` Alex Williamson
2019-04-15 18:53       ` Alex Williamson
2019-04-15 22:45       ` Jacob Pan
2019-04-15 22:45         ` Jacob Pan
2019-04-08 23:59   ` [PATCH 03/18] ioasid: Convert ioasid_idr to XArray Jacob Pan
2019-04-08 23:59     ` Jacob Pan
2019-04-08 23:59   ` [PATCH 04/18] driver core: add per device iommu param Jacob Pan
2019-04-08 23:59     ` Jacob Pan
2019-04-08 23:59   ` [PATCH 07/18] iommu: Introduce attach/detach_pasid_table API Jacob Pan
2019-04-08 23:59     ` Jacob Pan
2019-04-08 23:59   ` [PATCH 08/18] iommu: Introduce cache_invalidate API Jacob Pan
2019-04-08 23:59     ` Jacob Pan
2019-04-09 10:07     ` Andriy Shevchenko
2019-04-09 10:07       ` Andriy Shevchenko
2019-04-09 16:43       ` Jacob Pan
2019-04-09 16:43         ` Jacob Pan
2019-04-09 17:37         ` Andriy Shevchenko
2019-04-09 17:37           ` Andriy Shevchenko
2019-04-10 21:21           ` Jacob Pan
2019-04-10 21:21             ` Jacob Pan
2019-04-11 10:02             ` Andriy Shevchenko
2019-04-11 10:02               ` Andriy Shevchenko
2019-04-08 23:59   ` [PATCH 09/18] iommu/vt-d: Enlightened PASID allocation Jacob Pan
2019-04-08 23:59     ` Jacob Pan
     [not found]     ` <1554767973-30125-10-git-send-email-jacob.jun.pan-VuQAYsv1563Yd54FQh9/CA@public.gmane.org>
2019-04-09 10:08       ` Andriy Shevchenko
2019-04-09 10:08         ` Andriy Shevchenko
2019-04-09 16:34         ` Jacob Pan
2019-04-09 16:34           ` Jacob Pan
2019-04-15 17:25   ` [PATCH 00/18] Shared virtual address IOMMU and VT-d support Jacob Pan
2019-04-15 17:25     ` Jacob Pan

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=20190415143711.13c29308@x1.home \
    --to=alex.williamson-h+wxahxf7alqt0dzr+alfa@public.gmane.org \
    --cc=andriy.shevchenko-VuQAYsv1563Yd54FQh9/CA@public.gmane.org \
    --cc=ashok.raj-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org \
    --cc=dwmw2-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org \
    --cc=iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org \
    --cc=jacob.jun.pan-VuQAYsv1563Yd54FQh9/CA@public.gmane.org \
    --cc=jean-philippe.brucker-5wv7dgnIgG8@public.gmane.org \
    --cc=kevin.tian-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org \
    --cc=linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.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