Merge: Rebase MANA hyperv driver to v6.14

MR: https://gitlab.com/redhat/centos-stream/src/kernel/centos-stream-9/-/merge_requests/6508

This rebases MANA hyperv driver to upstream kernel 6.14-rc5

JIRA: https://issues.redhat.com/browse/RHEL-80098

Tested: Smoke boot test on azure machine, IB portion not yet tested.

Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>

Approved-by: Kamal Heib <kheib@redhat.com>
Approved-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Approved-by: CKI KWF Bot <cki-ci-bot+kwf-gitlab-com@redhat.com>

Merged-by: Augusto Caringi <acaringi@redhat.com>

Author: caringi

drivers/infiniband/hw/mana/main.c

@@ -174,7 +174,7 @@ static int mana_gd_allocate_doorbell_page(struct gdma_context *gc,
 
 	req.resource_type = GDMA_RESOURCE_DOORBELL_PAGE;
 	req.num_resources = 1;
-	req.alignment = 1;
+	req.alignment = PAGE_SIZE / MANA_PAGE_SIZE;
 
 	/* Have GDMA start searching from 0 */
 	req.allocated_resources = 0;
@@ -383,7 +383,7 @@ static int mana_ib_gd_create_dma_region(struct mana_ib_dev *dev, struct ib_umem
 
 	create_req->length = umem->length;
 	create_req->offset_in_page = ib_umem_dma_offset(umem, page_sz);
-	create_req->gdma_page_type = order_base_2(page_sz) - PAGE_SHIFT;
+	create_req->gdma_page_type = order_base_2(page_sz) - MANA_PAGE_SHIFT;
 	create_req->page_count = num_pages_total;
 
 	ibdev_dbg(&dev->ib_dev, "size_dma_region %lu num_pages_total %lu\n",
@@ -511,13 +511,13 @@ int mana_ib_mmap(struct ib_ucontext *ibcontext, struct vm_area_struct *vma)
 	      PAGE_SHIFT;
 	prot = pgprot_writecombine(vma->vm_page_prot);
 
-	ret = rdma_user_mmap_io(ibcontext, vma, pfn, gc->db_page_size, prot,
+	ret = rdma_user_mmap_io(ibcontext, vma, pfn, PAGE_SIZE, prot,
 				NULL);
 	if (ret)
 		ibdev_dbg(ibdev, "can't rdma_user_mmap_io ret %d\n", ret);
 	else
-		ibdev_dbg(ibdev, "mapped I/O pfn 0x%llx page_size %u, ret %d\n",
-			  pfn, gc->db_page_size, ret);
+		ibdev_dbg(ibdev, "mapped I/O pfn 0x%llx page_size %lu, ret %d\n",
+			  pfn, PAGE_SIZE, ret);
 
 	return ret;
 }

drivers/net/ethernet/microsoft/mana/gdma_main.c

@@ -1,6 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
 /* Copyright (c) 2021, Microsoft Corporation. */
 
+#include <linux/debugfs.h>
 #include <linux/module.h>
 #include <linux/pci.h>
 #include <linux/utsname.h>
@@ -9,6 +10,7 @@
 #include <net/mana/mana.h>
 
 #include <linux/cpu.h>
+struct dentry *mana_debugfs_root;
 
 static u32 mana_gd_r32(struct gdma_context *g, u64 offset)
 {
@@ -1317,7 +1319,7 @@ static int mana_gd_setup_irqs(struct pci_dev *pdev)
 				   GFP_KERNEL);
 	if (!gc->irq_contexts) {
 		err = -ENOMEM;
-		goto free_irq_vector;
+		goto free_irq_array;
 	}
 
 	for (i = 0; i < nvec; i++) {
@@ -1374,6 +1376,7 @@ static int mana_gd_setup_irqs(struct pci_dev *pdev)
 	gc->max_num_msix = nvec;
 	gc->num_msix_usable = nvec;
 	cpus_read_unlock();
+	kfree(irqs);
 	return 0;
 
 free_irq:
@@ -1386,8 +1389,9 @@ static int mana_gd_setup_irqs(struct pci_dev *pdev)
 	}
 
 	kfree(gc->irq_contexts);
-	kfree(irqs);
 	gc->irq_contexts = NULL;
+free_irq_array:
+	kfree(irqs);
 free_irq_vector:
 	cpus_read_unlock();
 	pci_free_irq_vectors(pdev);
@@ -1518,6 +1522,12 @@ static int mana_gd_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	gc->bar0_va = bar0_va;
 	gc->dev = &pdev->dev;
 
+	if (gc->is_pf)
+		gc->mana_pci_debugfs = debugfs_create_dir("0", mana_debugfs_root);
+	else
+		gc->mana_pci_debugfs = debugfs_create_dir(pci_slot_name(pdev->slot),
+							  mana_debugfs_root);
+
 	err = mana_gd_setup(pdev);
 	if (err)
 		goto unmap_bar;
@@ -1531,6 +1541,13 @@ static int mana_gd_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 cleanup_gd:
 	mana_gd_cleanup(pdev);
 unmap_bar:
+	/*
+	 * at this point we know that the other debugfs child dir/files
+	 * are either not yet created or are already cleaned up.
+	 * The pci debugfs folder clean-up now, will only be cleaning up
+	 * adapter-MTU file and apc->mana_pci_debugfs folder.
+	 */
+	debugfs_remove_recursive(gc->mana_pci_debugfs);
 	pci_iounmap(pdev, bar0_va);
 free_gc:
 	pci_set_drvdata(pdev, NULL);
@@ -1551,6 +1568,8 @@ static void mana_gd_remove(struct pci_dev *pdev)
 
 	mana_gd_cleanup(pdev);
 
+	debugfs_remove_recursive(gc->mana_pci_debugfs);
+
 	pci_iounmap(pdev, gc->bar0_va);
 
 	vfree(gc);
@@ -1602,6 +1621,8 @@ static void mana_gd_shutdown(struct pci_dev *pdev)
 
 	mana_gd_cleanup(pdev);
 
+	debugfs_remove_recursive(gc->mana_pci_debugfs);
+
 	pci_disable_device(pdev);
 }
 
@@ -1621,7 +1642,28 @@ static struct pci_driver mana_driver = {
 	.shutdown	= mana_gd_shutdown,
 };
 
-module_pci_driver(mana_driver);
+static int __init mana_driver_init(void)
+{
+	int err;
+
+	mana_debugfs_root = debugfs_create_dir("mana", NULL);
+
+	err = pci_register_driver(&mana_driver);
+	if (err)
+		debugfs_remove(mana_debugfs_root);
+
+	return err;
+}
+
+static void __exit mana_driver_exit(void)
+{
+	pci_unregister_driver(&mana_driver);
+
+	debugfs_remove(mana_debugfs_root);
+}
+
+module_init(mana_driver_init);
+module_exit(mana_driver_exit);
 
 MODULE_DEVICE_TABLE(pci, mana_id_table);
 

drivers/net/ethernet/microsoft/mana/hw_channel.c

@@ -51,16 +51,41 @@ static int mana_hwc_verify_resp_msg(const struct hwc_caller_ctx *caller_ctx,
 	return 0;
 }
 
+static int mana_hwc_post_rx_wqe(const struct hwc_wq *hwc_rxq,
+				struct hwc_work_request *req)
+{
+	struct device *dev = hwc_rxq->hwc->dev;
+	struct gdma_sge *sge;
+	int err;
+
+	sge = &req->sge;
+	sge->address = (u64)req->buf_sge_addr;
+	sge->mem_key = hwc_rxq->msg_buf->gpa_mkey;
+	sge->size = req->buf_len;
+
+	memset(&req->wqe_req, 0, sizeof(struct gdma_wqe_request));
+	req->wqe_req.sgl = sge;
+	req->wqe_req.num_sge = 1;
+	req->wqe_req.client_data_unit = 0;
+
+	err = mana_gd_post_and_ring(hwc_rxq->gdma_wq, &req->wqe_req, NULL);
+	if (err)
+		dev_err(dev, "Failed to post WQE on HWC RQ: %d\n", err);
+	return err;
+}
+
 static void mana_hwc_handle_resp(struct hw_channel_context *hwc, u32 resp_len,
-				 const struct gdma_resp_hdr *resp_msg)
+				 struct hwc_work_request *rx_req)
 {
+	const struct gdma_resp_hdr *resp_msg = rx_req->buf_va;
 	struct hwc_caller_ctx *ctx;
 	int err;
 
 	if (!test_bit(resp_msg->response.hwc_msg_id,
 		      hwc->inflight_msg_res.map)) {
 		dev_err(hwc->dev, "hwc_rx: invalid msg_id = %u\n",
 			resp_msg->response.hwc_msg_id);
+		mana_hwc_post_rx_wqe(hwc->rxq, rx_req);
 		return;
 	}
 
@@ -74,30 +99,13 @@ static void mana_hwc_handle_resp(struct hw_channel_context *hwc, u32 resp_len,
 	memcpy(ctx->output_buf, resp_msg, resp_len);
 out:
 	ctx->error = err;
-	complete(&ctx->comp_event);
-}
-
-static int mana_hwc_post_rx_wqe(const struct hwc_wq *hwc_rxq,
-				struct hwc_work_request *req)
-{
-	struct device *dev = hwc_rxq->hwc->dev;
-	struct gdma_sge *sge;
-	int err;
-
-	sge = &req->sge;
-	sge->address = (u64)req->buf_sge_addr;
-	sge->mem_key = hwc_rxq->msg_buf->gpa_mkey;
-	sge->size = req->buf_len;
 
-	memset(&req->wqe_req, 0, sizeof(struct gdma_wqe_request));
-	req->wqe_req.sgl = sge;
-	req->wqe_req.num_sge = 1;
-	req->wqe_req.client_data_unit = 0;
+	/* Must post rx wqe before complete(), otherwise the next rx may
+	 * hit no_wqe error.
+	 */
+	mana_hwc_post_rx_wqe(hwc->rxq, rx_req);
 
-	err = mana_gd_post_and_ring(hwc_rxq->gdma_wq, &req->wqe_req, NULL);
-	if (err)
-		dev_err(dev, "Failed to post WQE on HWC RQ: %d\n", err);
-	return err;
+	complete(&ctx->comp_event);
 }
 
 static void mana_hwc_init_event_handler(void *ctx, struct gdma_queue *q_self,
@@ -234,14 +242,12 @@ static void mana_hwc_rx_event_handler(void *ctx, u32 gdma_rxq_id,
 		return;
 	}
 
-	mana_hwc_handle_resp(hwc, rx_oob->tx_oob_data_size, resp);
+	mana_hwc_handle_resp(hwc, rx_oob->tx_oob_data_size, rx_req);
 
-	/* Do no longer use 'resp', because the buffer is posted to the HW
-	 * in the below mana_hwc_post_rx_wqe().
+	/* Can no longer use 'resp', because the buffer is posted to the HW
+	 * in mana_hwc_handle_resp() above.
 	 */
 	resp = NULL;
-
-	mana_hwc_post_rx_wqe(hwc_rxq, rx_req);
 }
 
 static void mana_hwc_tx_event_handler(void *ctx, u32 gdma_txq_id,