From: Santosh Shukla <sshukla@mvista.com>
To: "Burakov, Anatoly" <anatoly.burakov@intel.com>
Cc: "dev@dpdk.org" <dev@dpdk.org>
Subject: Re: [dpdk-dev] [PATCH v5 08/11] eal: pci: introduce RTE_KDRV_VFIO_NOIOMMUi driver mode
Date: Wed, 20 Jan 2016 00:06:05 +0530 [thread overview]
Message-ID: <CAAyOgsaL6jNcoGpb3uswP78+JHE9K=+USWe4kBxGeQgUCtZwKg@mail.gmail.com> (raw)
In-Reply-To: <C6ECDF3AB251BE4894318F4E45123697820BE4F2@IRSMSX109.ger.corp.intel.com>
On Tue, Jan 19, 2016 at 7:48 PM, Burakov, Anatoly
<anatoly.burakov@intel.com> wrote:
> Hi Santosh,
>
>> +int
>> +pci_vfio_is_noiommu(struct rte_pci_device *pci_dev) {
>> + FILE *fp;
>> + struct rte_pci_addr *loc;
>> + const char *path =
>> "/sys/module/vfio/parameters/enable_unsafe_noiommu_mode";
>> + char filename[PATH_MAX] = {0};
>> + char buf[PATH_MAX] = {0};
>> +
>> + /*
>> + * 1. chk vfio-noiommu mode set in kernel driver
>> + * 2. verify pci device attached to vfio-noiommu driver
>> + * example:
>> + * cd /sys/bus/pci/drivers/vfio-pci/<virtio_dev_addr>/iommu_group
>> + * > cat name
>> + * > vfio-noiommu --> means virtio_dev attached to vfio-noiommu
>> driver
>> + */
>> +
>> + fp = fopen(path, "r");
>> + if (fp == NULL) {
>> + RTE_LOG(ERR, EAL, "can't open %s\n", path);
>> + return -1;
>> + }
>> +
>> + if (fread(buf, sizeof(char), 1, fp) != 1) {
>> + RTE_LOG(ERR, EAL, "can't read from file %s\n", path);
>> + fclose(fp);
>> + return -1;
>> + }
>> +
>> + if (strncmp(buf, "Y", 1) != 0) {
>> + RTE_LOG(ERR, EAL, "[%s]: vfio: noiommu mode not set\n",
>> path);
>> + fclose(fp);
>> + return -1;
>> + }
>> +
>> + fclose(fp);
>> +
>> + /* 2. chk whether attached driver is vfio-noiommu or not */
>> + loc = &pci_dev->addr;
>> + snprintf(filename, sizeof(filename),
>> + SYSFS_PCI_DEVICES "/" PCI_PRI_FMT
>> "/iommu_group/name",
>> + loc->domain, loc->bus, loc->devid, loc->function);
>> +
>> + /* check for vfio-noiommu */
>> + fp = fopen(filename, "r");
>> + if (fp == NULL) {
>> + RTE_LOG(ERR, EAL, "can't open %s\n", filename);
>> + return -1;
>> + }
>> +
>> + if (fread(buf, sizeof(char), sizeof("vfio-noiommu"), fp) !=
>> + sizeof("vfio-noiommu")) {
>> + RTE_LOG(ERR, EAL, "can't read from file %s\n", filename);
>> + fclose(fp);
>> + return -1;
>> + }
>> +
>> + if (strncmp(buf, "vfio-noiommu", strlen("vfio-noiommu")) != 0) {
>> + RTE_LOG(ERR, EAL, "not a vfio-noiommu driver\n");
>> + fclose(fp);
>> + return -1;
>> + }
>> +
>> + fclose(fp);
>> +
>> + return 0;
>> +}
>
> Since this is a public non-performance critical API, shouldn't we check if pci_dev is NULL? Otherwise the patch-set seems fine to me as far as VFIO parts are concerned.
>
pci_scan_one() uses this api for now and it populate pci_dev before
pci_vfio_is_noiommu() could use. So didn't though to add a check, But
you are right in case any other module want to use this api. Sending
patch now. Thanks.
> Thanks,
> Anatoly
next prev parent reply other threads:[~2016-01-19 18:36 UTC|newest]
Thread overview: 38+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-01-19 11:46 [dpdk-dev] [PATCH v5 00/11] Add virtio support for arm/arm64 Santosh Shukla
2016-01-19 11:46 ` [dpdk-dev] [PATCH v5 01/11] virtio: Introduce config RTE_VIRTIO_INC_VECTOR Santosh Shukla
2016-01-27 2:23 ` Santosh Shukla
2016-01-27 2:33 ` Yuanhan Liu
2016-01-29 4:32 ` Santosh Shukla
2016-01-29 4:42 ` Yuanhan Liu
2016-01-29 4:45 ` Santosh Shukla
2016-01-19 11:46 ` [dpdk-dev] [PATCH v5 02/11] linuxapp: eal: arm: Always return 0 for rte_eal_iopl_init() Santosh Shukla
2016-01-21 9:41 ` David Marchand
2016-01-21 10:07 ` Santosh Shukla
2016-01-19 11:46 ` [dpdk-dev] [PATCH v5 03/11] linuxapp/vfio: ignore mapping for ioport region Santosh Shukla
2016-01-27 2:24 ` Santosh Shukla
2016-01-19 11:46 ` [dpdk-dev] [PATCH v5 04/11] virtio_pci.h: build fix for sys/io.h for non-x86 arch Santosh Shukla
2016-01-27 2:25 ` Santosh Shukla
2016-01-19 11:46 ` [dpdk-dev] [PATCH v5 05/11] eal: pci: vfio: add rd/wr func for pci bar space Santosh Shukla
2016-01-21 9:42 ` David Marchand
2016-01-21 10:08 ` Santosh Shukla
2016-01-19 11:46 ` [dpdk-dev] [PATCH v5 06/11] virtio: vfio: add api support to rd/wr ioport bar Santosh Shukla
2016-01-29 7:07 ` Yuanhan Liu
2016-01-29 7:16 ` Santosh Shukla
2016-01-19 11:46 ` [dpdk-dev] [PATCH v5 07/11] virtio: pci: extend virtio pci rw api for vfio interface Santosh Shukla
2016-01-19 11:46 ` [dpdk-dev] [PATCH v5 08/11] eal: pci: introduce RTE_KDRV_VFIO_NOIOMMUi driver mode Santosh Shukla
2016-01-19 14:18 ` Burakov, Anatoly
2016-01-19 18:36 ` Santosh Shukla [this message]
2016-01-19 11:46 ` [dpdk-dev] [PATCH v5 09/11] virtio_pci: do not parse if interface is vfio-noiommu Santosh Shukla
2016-01-29 7:17 ` Yuanhan Liu
2016-01-29 7:22 ` Santosh Shukla
2016-01-29 7:34 ` Yuanhan Liu
2016-01-29 9:02 ` Thomas Monjalon
2016-01-29 9:14 ` Yuanhan Liu
2016-01-29 9:16 ` Santosh Shukla
2016-01-19 11:46 ` [dpdk-dev] [PATCH v5 10/11] virtio: pci: add dummy func definition for in/outb for non-x86 arch Santosh Shukla
2016-01-27 10:37 ` Santosh Shukla
2016-01-29 7:01 ` Yuanhan Liu
2016-01-29 7:31 ` Santosh Shukla
2016-01-29 7:38 ` Yuanhan Liu
2016-01-19 11:46 ` [dpdk-dev] [PATCH v5 11/11] vfio: Support for no-IOMMU mode Santosh Shukla
2016-01-25 7:25 ` [dpdk-dev] [PATCH v5 00/11] Add virtio support for arm/arm64 Santosh Shukla
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='CAAyOgsaL6jNcoGpb3uswP78+JHE9K=+USWe4kBxGeQgUCtZwKg@mail.gmail.com' \
--to=sshukla@mvista.com \
--cc=anatoly.burakov@intel.com \
--cc=dev@dpdk.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;
as well as URLs for NNTP newsgroup(s).