patches for DPDK stable branches
 help / color / mirror / Atom feed
From: Abdullah Sevincer <abdullah.sevincer@intel.com>
To: stable@dpdk.org
Cc: Abdullah Sevincer <abdullah.sevincer@intel.com>
Subject: [PATCH 22.11] event/dlb2: fix disable PASID
Date: Thu,  9 Nov 2023 13:45:44 -0600	[thread overview]
Message-ID: <20231109194544.3246038-1-abdullah.sevincer@intel.com> (raw)

[ upstream commit 5a6878335b8179337ec2d9931debf1f46525e8fc ]

In vfio-pci driver when PASID is enabled by default DLB hardware puts
DLB in SIOV mode. This breaks DLB PF-PMD mode. For DLB PF-PMD mode to
function properly PASID needs to be disabled.

In this commit this issue is addressed and PASID is disabled by writing
a zero to PASID control register.

Fixes: 5433956d5185 ("event/dlb2: add eventdev probe")

Signed-off-by: Abdullah Sevincer <abdullah.sevincer@intel.com>
---
 drivers/event/dlb2/pf/dlb2_main.c | 27 +++++++++++++++++++++++++++
 1 file changed, 27 insertions(+)

diff --git a/drivers/event/dlb2/pf/dlb2_main.c b/drivers/event/dlb2/pf/dlb2_main.c
index 717aa4fc08..63868e2388 100644
--- a/drivers/event/dlb2/pf/dlb2_main.c
+++ b/drivers/event/dlb2/pf/dlb2_main.c
@@ -46,6 +46,7 @@
 #define DLB2_PCI_CAP_ID_MSIX      0x11
 #define DLB2_PCI_EXT_CAP_ID_PRI   0x13
 #define DLB2_PCI_EXT_CAP_ID_ACS   0xD
+#define DLB2_PCI_EXT_CAP_ID_PASID 0x1B	/* Process Address Space ID */
 
 #define DLB2_PCI_PRI_CTRL_ENABLE         0x1
 #define DLB2_PCI_PRI_ALLOC_REQ           0xC
@@ -64,6 +65,8 @@
 #define DLB2_PCI_ACS_CR                  0x8
 #define DLB2_PCI_ACS_UF                  0x10
 #define DLB2_PCI_ACS_EC                  0x20
+#define DLB2_PCI_PASID_CTRL              0x06    /* PASID control register */
+#define DLB2_PCI_PASID_CAP_OFFSET        0x148   /* PASID capability offset */
 
 static int dlb2_pci_find_capability(struct rte_pci_device *pdev, uint32_t id)
 {
@@ -257,12 +260,14 @@ dlb2_pf_reset(struct dlb2_dev *dlb2_dev)
 	uint16_t rt_ctl_word;
 	uint32_t pri_reqs_dword;
 	uint16_t pri_ctrl_word;
+	uint16_t pasid_ctrl;
 
 	int pcie_cap_offset;
 	int pri_cap_offset;
 	int msix_cap_offset;
 	int err_cap_offset;
 	int acs_cap_offset;
+	int pasid_cap_offset;
 	int wait_count;
 
 	uint16_t devsta_busy_word;
@@ -582,6 +587,28 @@ dlb2_pf_reset(struct dlb2_dev *dlb2_dev)
 		}
 	}
 
+	/* The current Linux kernel vfio driver does not expose PASID capability to
+	 * users. It also enables PASID by default, which breaks DLB PF PMD. We have
+	 * to use the hardcoded offset for now to disable PASID.
+	 */
+	pasid_cap_offset = DLB2_PCI_PASID_CAP_OFFSET;
+
+	off = pasid_cap_offset + DLB2_PCI_PASID_CTRL;
+	if (rte_pci_read_config(pdev, &pasid_ctrl, 2, off) != 2)
+		pasid_ctrl = 0;
+
+	if (pasid_ctrl) {
+		DLB2_INFO(dlb2_dev, "DLB2 disabling pasid...\n");
+
+		pasid_ctrl = 0;
+		ret = rte_pci_write_config(pdev, &pasid_ctrl, 2, off);
+		if (ret != 2) {
+			DLB2_LOG_ERR("[%s()] failed to write the pcie config space at offset %d\n",
+				__func__, (int)off);
+			return ret;
+		}
+	}
+
 	return 0;
 }
 
-- 
2.39.1.windows.1


             reply	other threads:[~2023-11-09 19:46 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-11-09 19:45 Abdullah Sevincer [this message]
2023-11-09 21:42 ` Luca Boccassi
2023-11-23 10:47 ` Kevin Traynor
2023-12-11  6:25 ` Xueming(Steven) Li

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=20231109194544.3246038-1-abdullah.sevincer@intel.com \
    --to=abdullah.sevincer@intel.com \
    --cc=stable@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).