linuxppc-dev.lists.ozlabs.org archive mirror
 help / color / mirror / Atom feed
From: "Ian Munsie" <imunsie@au1.ibm.com>
To: Michael Ellerman <mpe@ellerman.id.au>,
	Michael Neuling <mikey@neuling.org>,
	Frederic Barrat <fbarrat@linux.vnet.ibm.com>,
	Andrew Donnellan <andrew.donnellan@au1.ibm.com>,
	linuxppc-dev@lists.ozlabs.org, Huy Nguyen <huyn@mellanox.com>
Cc: Ian Munsie <imunsie@au1.ibm.com>
Subject: [PATCH 09/14] cxl: Add preliminary workaround for CX4 interrupt limitation
Date: Mon,  4 Jul 2016 23:22:07 +1000	[thread overview]
Message-ID: <1467638532-9250-10-git-send-email-imunsie@au.ibm.com> (raw)
In-Reply-To: <1467638532-9250-1-git-send-email-imunsie@au.ibm.com>

From: Ian Munsie <imunsie@au1.ibm.com>

The Mellanox CX4 has a hardware limitation where only 4 bits of the
AFU interrupt number can be passed to the XSL when sending an interrupt,
limiting it to only 15 interrupts per context (AFU interrupt number 0 is
invalid).

In order to overcome this, we will allocate additional contexts linked
to the default context as extra address space for the extra interrupts -
this will be implemented in the next patch.

This patch adds the preliminary support to allow this, by way of adding
a linked list in the context structure that we use to keep track of the
contexts dedicated to interrupts, and an API to simultaneously iterate
over the related context structures, AFU interrupt numbers and hardware
interrupt numbers. The point of using a single API to iterate these is
to hide some of the details of the iteration from external code, and to
reduce the number of APIs that need to be exported via base.c to allow
built in code to call.

Signed-off-by: Ian Munsie <imunsie@au1.ibm.com>
---
 drivers/misc/cxl/api.c     | 15 +++++++++++++++
 drivers/misc/cxl/base.c    | 17 +++++++++++++++++
 drivers/misc/cxl/context.c |  1 +
 drivers/misc/cxl/cxl.h     | 10 ++++++++++
 drivers/misc/cxl/main.c    |  1 +
 include/misc/cxl.h         |  9 +++++++++
 6 files changed, 53 insertions(+)

diff --git a/drivers/misc/cxl/api.c b/drivers/misc/cxl/api.c
index 1e2c0d9..f02a859 100644
--- a/drivers/misc/cxl/api.c
+++ b/drivers/misc/cxl/api.c
@@ -97,6 +97,21 @@ static irq_hw_number_t cxl_find_afu_irq(struct cxl_context *ctx, int num)
 	return 0;
 }
 
+int _cxl_next_msi_hwirq(struct pci_dev *pdev, struct cxl_context **ctx, int *afu_irq)
+{
+	if (*ctx == NULL || *afu_irq == 0) {
+		*afu_irq = 1;
+		*ctx = cxl_get_context(pdev);
+	} else {
+		(*afu_irq)++;
+		if (*afu_irq > cxl_get_max_irqs_per_process(pdev)) {
+			*ctx = list_next_entry(*ctx, extra_irq_contexts);
+			*afu_irq = 1;
+		}
+	}
+	return cxl_find_afu_irq(*ctx, *afu_irq);
+}
+/* Exported via cxl_base */
 
 int cxl_set_priv(struct cxl_context *ctx, void *priv)
 {
diff --git a/drivers/misc/cxl/base.c b/drivers/misc/cxl/base.c
index af20b34..0f89ea9 100644
--- a/drivers/misc/cxl/base.c
+++ b/drivers/misc/cxl/base.c
@@ -141,6 +141,23 @@ void cxl_pci_disable_device(struct pci_dev *dev)
 }
 EXPORT_SYMBOL_GPL(cxl_pci_disable_device);
 
+int cxl_next_msi_hwirq(struct pci_dev *pdev, struct cxl_context **ctx, int *afu_irq)
+{
+	int ret;
+	struct cxl_calls *calls;
+
+	calls = cxl_calls_get();
+	if (!calls)
+		return -EBUSY;
+
+	ret = calls->cxl_next_msi_hwirq(pdev, ctx, afu_irq);
+
+	cxl_calls_put(calls);
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(cxl_next_msi_hwirq);
+
 static int __init cxl_base_init(void)
 {
 	struct device_node *np = NULL;
diff --git a/drivers/misc/cxl/context.c b/drivers/misc/cxl/context.c
index edbb99e..2616cddb 100644
--- a/drivers/misc/cxl/context.c
+++ b/drivers/misc/cxl/context.c
@@ -68,6 +68,7 @@ int cxl_context_init(struct cxl_context *ctx, struct cxl_afu *afu, bool master,
 	ctx->pending_afu_err = false;
 
 	INIT_LIST_HEAD(&ctx->irq_names);
+	INIT_LIST_HEAD(&ctx->extra_irq_contexts);
 
 	/*
 	 * When we have to destroy all contexts in cxl_context_detach_all() we
diff --git a/drivers/misc/cxl/cxl.h b/drivers/misc/cxl/cxl.h
index c94b54f..67464c9 100644
--- a/drivers/misc/cxl/cxl.h
+++ b/drivers/misc/cxl/cxl.h
@@ -525,6 +525,14 @@ struct cxl_context {
 	atomic_t afu_driver_events;
 
 	struct rcu_head rcu;
+
+	/*
+	 * Only used when more interrupts are allocated via
+	 * pci_enable_msix_range than are supported in the default context, to
+	 * use additional contexts to overcome the limitation. i.e. Mellanox
+	 * CX4 only:
+	 */
+	struct list_head extra_irq_contexts;
 };
 
 struct cxl_service_layer_ops {
@@ -710,11 +718,13 @@ ssize_t cxl_pci_afu_read_err_buffer(struct cxl_afu *afu, char *buf,
 /* Internal functions wrapped in cxl_base to allow PHB to call them */
 bool _cxl_pci_associate_default_context(struct pci_dev *dev, struct cxl_afu *afu);
 void _cxl_pci_disable_device(struct pci_dev *dev);
+int _cxl_next_msi_hwirq(struct pci_dev *pdev, struct cxl_context **ctx, int *afu_irq);
 
 struct cxl_calls {
 	void (*cxl_slbia)(struct mm_struct *mm);
 	bool (*cxl_pci_associate_default_context)(struct pci_dev *dev, struct cxl_afu *afu);
 	void (*cxl_pci_disable_device)(struct pci_dev *dev);
+	int (*cxl_next_msi_hwirq)(struct pci_dev *pdev, struct cxl_context **ctx, int *afu_irq);
 
 	struct module *owner;
 };
diff --git a/drivers/misc/cxl/main.c b/drivers/misc/cxl/main.c
index 4e5474b..66fac71 100644
--- a/drivers/misc/cxl/main.c
+++ b/drivers/misc/cxl/main.c
@@ -112,6 +112,7 @@ static struct cxl_calls cxl_calls = {
 	.cxl_slbia = cxl_slbia_core,
 	.cxl_pci_associate_default_context = _cxl_pci_associate_default_context,
 	.cxl_pci_disable_device = _cxl_pci_disable_device,
+	.cxl_next_msi_hwirq = _cxl_next_msi_hwirq,
 	.owner = THIS_MODULE,
 };
 
diff --git a/include/misc/cxl.h b/include/misc/cxl.h
index fc07ed4..ed81a17 100644
--- a/include/misc/cxl.h
+++ b/include/misc/cxl.h
@@ -178,6 +178,15 @@ int cxl_set_max_irqs_per_process(struct pci_dev *dev, int irqs);
 int cxl_get_max_irqs_per_process(struct pci_dev *dev);
 
 /*
+ * Use to simultaneously iterate over hardware interrupt numbers, contexts and
+ * afu interrupt numbers allocated for the device via pci_enable_msix_range and
+ * is a useful convinience function when working with hardware that has
+ * limitations on the number of interrupts per process. *ctx and *afu_irq
+ * should be NULL and 0 to start the iteration.
+ */
+int cxl_next_msi_hwirq(struct pci_dev *pdev, struct cxl_context **ctx, int *afu_irq);
+
+/*
  * These calls allow drivers to create their own file descriptors and make them
  * identical to the cxl file descriptor user API. An example use case:
  *
-- 
2.8.1

  parent reply	other threads:[~2016-07-04 13:22 UTC|newest]

Thread overview: 46+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-07-04 13:21 powerpc / cxl: Add support for the Mellanox CX4 in cxl mode Ian Munsie
2016-07-04 13:21 ` [PATCH 01/14] powerpc/powernv: Split cxl code out into a separate file Ian Munsie
2016-07-06  3:44   ` Andrew Donnellan
2016-07-06 16:27   ` Frederic Barrat
2016-07-04 13:22 ` [PATCH 02/14] cxl: Add cxl_slot_is_supported API Ian Munsie
2016-07-06  2:02   ` Andrew Donnellan
2016-07-06 16:36   ` Frederic Barrat
2016-07-04 13:22 ` [PATCH 03/14] cxl: Enable bus mastering for devices using CAPP DMA mode Ian Munsie
2016-07-06  4:04   ` Andrew Donnellan
2016-07-06 16:37   ` Frederic Barrat
2016-07-04 13:22 ` [PATCH 04/14] cxl: Move cxl_afu_get / cxl_afu_put to base Ian Munsie
2016-07-05  2:10   ` Andrew Donnellan
2016-07-06 16:45   ` Frederic Barrat
2016-07-04 13:22 ` [PATCH 05/14] cxl: Allow a default context to be associated with an external pci_dev Ian Munsie
2016-07-06 16:51   ` Frederic Barrat
2016-07-04 13:22 ` [PATCH 06/14] powerpc/powernv: Add support for the cxl kernel api on the real phb Ian Munsie
2016-07-06 17:38   ` Frederic Barrat
2016-07-07  6:28     ` Ian Munsie
2016-07-04 13:22 ` [PATCH 07/14] cxl: Add support for using the kernel API with a real PHB Ian Munsie
2016-07-06 17:39   ` Frederic Barrat
2016-07-06 18:30   ` Frederic Barrat
2016-07-07  6:32     ` Ian Munsie
2016-07-04 13:22 ` [PATCH 08/14] cxl: Add kernel APIs to get & set the max irqs per context Ian Munsie
2016-07-06 18:11   ` Frederic Barrat
2016-07-07  6:00     ` Ian Munsie
2016-07-04 13:22 ` Ian Munsie [this message]
2016-07-06 18:34   ` [PATCH 09/14] cxl: Add preliminary workaround for CX4 interrupt limitation Frederic Barrat
2016-07-04 13:22 ` [PATCH 10/14] cxl: Add support for interrupts on the Mellanox CX4 Ian Munsie
2016-07-06 18:41   ` Frederic Barrat
2016-07-07  6:03     ` Ian Munsie
2016-07-04 13:22 ` [PATCH 11/14] cxl: Workaround PE=0 hardware limitation in " Ian Munsie
2016-07-06  4:42   ` Andrew Donnellan
2016-07-06 18:42   ` Frederic Barrat
2016-07-04 13:22 ` [PATCH 12/14] PCI/hotplug: pnv_php: export symbols and move struct types needed by cxl Ian Munsie
2016-07-05  0:03   ` Gavin Shan
2016-07-05  1:08     ` Andrew Donnellan
2016-07-04 13:22 ` [PATCH 13/14] PCI/hotplug: pnv_php: handle OPAL_PCI_SLOT_OFFLINE power state Ian Munsie
2016-07-04 13:22 ` [PATCH 14/14] cxl: Add cxl_check_and_switch_mode() API to switch bi-modal cards Ian Munsie
2016-07-06  3:55   ` Andrew Donnellan
2016-07-06 18:51   ` Frederic Barrat
2016-07-07  1:18     ` Andrew Donnellan
2016-07-07  6:26       ` Ian Munsie
2016-07-07  6:44         ` Andrew Donnellan
2016-07-07  8:15           ` Andrew Donnellan
2016-07-11  9:19             ` Ian Munsie
2016-07-12  1:20               ` Andrew Donnellan

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=1467638532-9250-10-git-send-email-imunsie@au.ibm.com \
    --to=imunsie@au1.ibm.com \
    --cc=andrew.donnellan@au1.ibm.com \
    --cc=fbarrat@linux.vnet.ibm.com \
    --cc=huyn@mellanox.com \
    --cc=linuxppc-dev@lists.ozlabs.org \
    --cc=mikey@neuling.org \
    --cc=mpe@ellerman.id.au \
    /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).