Linux PCI subsystem development
 help / color / mirror / Atom feed
From: Dongdong Liu <liudongdong3@huawei.com>
To: Bjorn Helgaas <helgaas@kernel.org>
Cc: <hch@infradead.org>, <kw@linux.com>, <logang@deltatee.com>,
	<leon@kernel.org>, <linux-pci@vger.kernel.org>,
	<rajur@chelsio.com>, <hverkuil-cisco@xs4all.nl>,
	<linux-media@vger.kernel.org>, <netdev@vger.kernel.org>
Subject: Re: [PATCH V10 6/8] PCI/P2PDMA: Add a 10-Bit Tag check in P2PDMA
Date: Thu, 28 Oct 2021 15:56:45 +0800	[thread overview]
Message-ID: <5773efae-8fb6-bd63-0486-dfcd14d3c8ae@huawei.com> (raw)
In-Reply-To: <20211027212048.GA252528@bhelgaas>



On 2021/10/28 5:20, Bjorn Helgaas wrote:
> On Sat, Oct 09, 2021 at 06:49:36PM +0800, Dongdong Liu wrote:
>> Add a 10-Bit Tag check in the P2PDMA code to ensure that a device with
>> 10-Bit Tag Requester doesn't interact with a device that does not
>> support 10-Bit Tag Completer.
>
> Shouldn't this also take into account Extended Tags (8 bits)?  I think
> the only tag size guaranteed to be supported is 5 bits.
As all PCIe completers are required to support 8-bit tags, seems no need
to take into account Extended Tags.
>
>> Before that happens, the kernel should emit a warning.
>
> The warning is nice, but the critical thing is that the P2PDMA mapping
> should fail so we don't attempt DMA in this situation.  I guess that's
> sort of what you're saying with "ensure that a device ... doesn't
> interact with a device ..."
Yes, that is.
>
>> "echo 0 > /sys/bus/pci/devices/.../10bit_tag" to disable 10-Bit Tag
>> Requester for PF device.
>>
>> "echo 0 > /sys/bus/pci/devices/.../sriov_vf_10bit_tag_ctl" to disable
>> 10-Bit Tag Requester for VF device.
>>
>> Signed-off-by: Dongdong Liu <liudongdong3@huawei.com>
>> Reviewed-by: Logan Gunthorpe <logang@deltatee.com>
>> ---
>>  drivers/pci/p2pdma.c | 48 ++++++++++++++++++++++++++++++++++++++++++++
>>  1 file changed, 48 insertions(+)
>>
>> diff --git a/drivers/pci/p2pdma.c b/drivers/pci/p2pdma.c
>> index 50cdde3e9a8b..804e390f4c22 100644
>> --- a/drivers/pci/p2pdma.c
>> +++ b/drivers/pci/p2pdma.c
>> @@ -19,6 +19,7 @@
>>  #include <linux/random.h>
>>  #include <linux/seq_buf.h>
>>  #include <linux/xarray.h>
>> +#include "pci.h"
>>
>>  enum pci_p2pdma_map_type {
>>  	PCI_P2PDMA_MAP_UNKNOWN = 0,
>> @@ -410,6 +411,50 @@ static unsigned long map_types_idx(struct pci_dev *client)
>>  		(client->bus->number << 8) | client->devfn;
>>  }
>>
>> +static bool pci_10bit_tags_unsupported(struct pci_dev *a,
>> +				       struct pci_dev *b,
>> +				       bool verbose)
>> +{
>> +	bool req;
>> +	bool comp;
>> +	u16 ctl;
>> +	const char *str = "10bit_tag";
>> +
>> +	if (a->is_virtfn) {
>> +#ifdef CONFIG_PCI_IOV
>> +		req = !!(a->physfn->sriov->ctrl &
>> +			 PCI_SRIOV_CTRL_VF_10BIT_TAG_REQ_EN);
>> +#endif
>> +	} else {
>> +		pcie_capability_read_word(a, PCI_EXP_DEVCTL2, &ctl);
>> +		req = !!(ctl & PCI_EXP_DEVCTL2_10BIT_TAG_REQ_EN);
>> +	}
>> +
>> +	comp = !!(b->devcap2 & PCI_EXP_DEVCAP2_10BIT_TAG_COMP);
>> +	/* 10-bit tags not enabled on requester */
>> +	if (!req)
>> +		return false;
>> +
>> +	 /* Completer can handle anything */
>> +	if (comp)
>> +		return false;
>> +
>> +	if (!verbose)
>> +		return true;
>> +
>> +	pci_warn(a, "cannot be used for peer-to-peer DMA as 10-Bit Tag Requester enable is set for this device, but peer device (%s) does not support the 10-Bit Tag Completer\n",
>> +		 pci_name(b));
>> +
>> +	if (a->is_virtfn)
>> +		str = "sriov_vf_10bit_tag_ctl";
>> +
>> +	pci_warn(a, "to disable 10-Bit Tag Requester for this device, echo 0 > /sys/bus/pci/devices/%s/%s\n",
>> +		 pci_name(a), str);
>> +
>> +	return true;
>> +}
>> +
>>  /*
>>   * Calculate the P2PDMA mapping type and distance between two PCI devices.
>>   *
>> @@ -532,6 +577,9 @@ calc_map_type_and_dist(struct pci_dev *provider, struct pci_dev *client,
>>  		map_type = PCI_P2PDMA_MAP_NOT_SUPPORTED;
>>  	}
>>  done:
>> +	if (pci_10bit_tags_unsupported(client, provider, verbose))
>> +		map_type = PCI_P2PDMA_MAP_NOT_SUPPORTED;
>> +
>>  	rcu_read_lock();
>>  	p2pdma = rcu_dereference(provider->p2pdma);
>>  	if (p2pdma)
>> --
>> 2.22.0
>>
> .
>

  reply	other threads:[~2021-10-28  7:56 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-10-09 10:49 [PATCH V10 0/8] PCI: Enable 10-Bit tag support for PCIe devices Dongdong Liu
2021-10-09 10:49 ` [PATCH V10 1/8] PCI: Use cached devcap in more places Dongdong Liu
2021-10-09 10:49 ` [PATCH V10 2/8] PCI: Cache Device Capabilities 2 Register Dongdong Liu
2021-10-09 10:49 ` [PATCH V10 3/8] PCI: Add 10-Bit Tag register definitions Dongdong Liu
2021-10-09 10:49 ` [PATCH V10 4/8] PCI/sysfs: Add a 10-Bit Tag sysfs file PCIe Endpoint devices Dongdong Liu
2021-10-27 22:28   ` Bjorn Helgaas
2021-10-28  7:44     ` Dongdong Liu
2021-10-28 17:24       ` Bjorn Helgaas
2021-10-29  7:16         ` Dongdong Liu
2021-10-09 10:49 ` [PATCH V10 5/8] PCI/IOV: Add 10-Bit Tag sysfs files for VF devices Dongdong Liu
2021-10-09 10:49 ` [PATCH V10 6/8] PCI/P2PDMA: Add a 10-Bit Tag check in P2PDMA Dongdong Liu
2021-10-27 21:20   ` Bjorn Helgaas
2021-10-28  7:56     ` Dongdong Liu [this message]
2021-10-27 23:11   ` Bjorn Helgaas
2021-10-27 23:41     ` Logan Gunthorpe
2021-10-28  1:39       ` Bjorn Helgaas
2021-10-28 15:56         ` Logan Gunthorpe
2021-10-09 10:49 ` [PATCH V10 7/8] PCI: Enable 10-Bit Tag support for PCIe Endpoint device Dongdong Liu
2021-10-09 10:49 ` [PATCH V10 8/8] PCI/IOV: Enable 10-Bit Tag support for PCIe VF devices Dongdong Liu

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=5773efae-8fb6-bd63-0486-dfcd14d3c8ae@huawei.com \
    --to=liudongdong3@huawei.com \
    --cc=hch@infradead.org \
    --cc=helgaas@kernel.org \
    --cc=hverkuil-cisco@xs4all.nl \
    --cc=kw@linux.com \
    --cc=leon@kernel.org \
    --cc=linux-media@vger.kernel.org \
    --cc=linux-pci@vger.kernel.org \
    --cc=logang@deltatee.com \
    --cc=netdev@vger.kernel.org \
    --cc=rajur@chelsio.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