lkml.org 
[lkml]   [2007]   [Mar]   [22]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
Subject[PATCH 2.6.20.3] Flush writes to MSI-X table
From
Date
Because both MSI-X interrupt messages and MSI-X table writes are posted,
it's possible for them to cross while in-flight. This results in
interrupts being received long after the kernel thinks they're disabled,
and in interrupts being sent to stale vectors after rebalancing.

This patch performs a read flush after writes to the MSI-X table for
enable/disable and rebalancing operations. Because this is an expensive
operation, we do not perform the read flush after mask/unmask
operations. Hardware which supports MSI-X typically also supports some
sort of interrupt moderation, so a read-flush is not necessary for
mask/unmask operations.

This patch has been validated with (unreleased) network hardware which
uses MSI-X.

Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>

diff -urpN -X dontdiff linux-2.6.20.3-clean/arch/i386/kernel/io_apic.c linux-2.6.20.3/arch/i386/kernel/io_apic.c
--- linux-2.6.20.3-clean/arch/i386/kernel/io_apic.c 2007-02-04 10:44:54.000000000 -0800
+++ linux-2.6.20.3/arch/i386/kernel/io_apic.c 2007-03-22 10:33:47.000000000 -0700
@@ -2597,6 +2597,8 @@ static void set_msi_irq_affinity(unsigne
*/
static struct irq_chip msi_chip = {
.name = "PCI-MSI",
+ .enable = enable_msi_irq,
+ .disable = disable_msi_irq,
.unmask = unmask_msi_irq,
.mask = mask_msi_irq,
.ack = ack_ioapic_irq,
diff -urpN -X dontdiff linux-2.6.20.3-clean/arch/ia64/kernel/msi_ia64.c linux-2.6.20.3/arch/ia64/kernel/msi_ia64.c
--- linux-2.6.20.3-clean/arch/ia64/kernel/msi_ia64.c 2007-02-04 10:44:54.000000000 -0800
+++ linux-2.6.20.3/arch/ia64/kernel/msi_ia64.c 2007-03-22 10:33:47.000000000 -0700
@@ -116,6 +116,8 @@ static int ia64_msi_retrigger_irq(unsign
*/
static struct irq_chip ia64_msi_chip = {
.name = "PCI-MSI",
+ .enable = enable_msi_irq,
+ .disable = disable_msi_irq,
.mask = mask_msi_irq,
.unmask = unmask_msi_irq,
.ack = ia64_ack_msi_irq,
diff -urpN -X dontdiff linux-2.6.20.3-clean/arch/ia64/sn/kernel/msi_sn.c linux-2.6.20.3/arch/ia64/sn/kernel/msi_sn.c
--- linux-2.6.20.3-clean/arch/ia64/sn/kernel/msi_sn.c 2007-02-04 10:44:54.000000000 -0800
+++ linux-2.6.20.3/arch/ia64/sn/kernel/msi_sn.c 2007-03-22 10:33:47.000000000 -0700
@@ -216,6 +216,8 @@ static int sn_msi_retrigger_irq(unsigned

static struct irq_chip sn_msi_chip = {
.name = "PCI-MSI",
+ .enable = enable_msi_irq,
+ .disable = disable_msi_irq,
.mask = mask_msi_irq,
.unmask = unmask_msi_irq,
.ack = sn_ack_msi_irq,
diff -urpN -X dontdiff linux-2.6.20.3-clean/arch/x86_64/kernel/io_apic.c linux-2.6.20.3/arch/x86_64/kernel/io_apic.c
--- linux-2.6.20.3-clean/arch/x86_64/kernel/io_apic.c 2007-02-04 10:44:54.000000000 -0800
+++ linux-2.6.20.3/arch/x86_64/kernel/io_apic.c 2007-03-22 10:36:03.000000000 -0700
@@ -1923,6 +1923,7 @@ static void set_msi_irq_affinity(unsigne

cpus_and(mask, tmp, CPU_MASK_ALL);

+ msix_flush_writes(irq);
vector = assign_irq_vector(irq, mask, &tmp);
if (vector < 0)
return;
@@ -1937,6 +1938,7 @@ static void set_msi_irq_affinity(unsigne
msg.address_lo |= MSI_ADDR_DEST_ID(dest);

write_msi_msg(irq, &msg);
+ msix_flush_writes(irq);
set_native_irq_info(irq, mask);
}
#endif /* CONFIG_SMP */
@@ -1947,6 +1949,8 @@ static void set_msi_irq_affinity(unsigne
*/
static struct irq_chip msi_chip = {
.name = "PCI-MSI",
+ .enable = enable_msi_irq,
+ .disable = disable_msi_irq,
.unmask = unmask_msi_irq,
.mask = mask_msi_irq,
.ack = ack_apic_edge,
diff -urpN -X dontdiff linux-2.6.20.3-clean/drivers/pci/msi.c linux-2.6.20.3/drivers/pci/msi.c
--- linux-2.6.20.3-clean/drivers/pci/msi.c 2007-02-04 10:44:54.000000000 -0800
+++ linux-2.6.20.3/drivers/pci/msi.c 2007-03-22 10:33:47.000000000 -0700
@@ -40,6 +40,29 @@ static int msi_cache_init(void)
return 0;
}

+void msix_flush_writes(unsigned int irq)
+{
+ struct msi_desc *entry;
+
+ entry = msi_desc[irq];
+ BUG_ON(!entry || !entry->dev);
+ switch (entry->msi_attrib.type) {
+ case PCI_CAP_ID_MSI:
+ /* nothing to do */
+ break;
+ case PCI_CAP_ID_MSIX:
+ {
+ int offset = entry->msi_attrib.entry_nr * PCI_MSIX_ENTRY_SIZE +
+ PCI_MSIX_ENTRY_VECTOR_CTRL_OFFSET;
+ readl(entry->mask_base + offset);
+ break;
+ }
+ default:
+ BUG();
+ break;
+ }
+}
+
static void msi_set_mask_bit(unsigned int irq, int flag)
{
struct msi_desc *entry;
@@ -161,6 +184,17 @@ void unmask_msi_irq(unsigned int irq)
msi_set_mask_bit(irq, 0);
}

+void disable_msi_irq(unsigned int irq)
+{
+ msi_set_mask_bit(irq, 1);
+ msix_flush_writes(irq);
+}
+
+void enable_msi_irq(unsigned int irq)
+{
+ msi_set_mask_bit(irq, 0);
+ msix_flush_writes(irq);
+}
static int msi_free_irq(struct pci_dev* dev, int irq);
static int msi_init(void)
{
diff -urpN -X dontdiff linux-2.6.20.3-clean/include/linux/msi.h linux-2.6.20.3/include/linux/msi.h
--- linux-2.6.20.3-clean/include/linux/msi.h 2007-02-04 10:44:54.000000000 -0800
+++ linux-2.6.20.3/include/linux/msi.h 2007-03-22 10:33:47.000000000 -0700
@@ -10,9 +10,12 @@ struct msi_msg {
/* Heper functions */
extern void mask_msi_irq(unsigned int irq);
extern void unmask_msi_irq(unsigned int irq);
+extern void disable_msi_irq(unsigned int irq);
+extern void enable_msi_irq(unsigned int irq);
extern void read_msi_msg(unsigned int irq, struct msi_msg *msg);

extern void write_msi_msg(unsigned int irq, struct msi_msg *msg);
+extern void msix_flush_writes(unsigned int irq);

struct msi_desc {
struct {
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
\
 
 \ /
  Last update: 2007-03-22 22:11    [W:0.201 / U:0.332 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site