[prev in list] [next in list] [prev in thread] [next in thread]
List: kvm
Subject: [kvm-unit-tests PATCH 6/8] intel-iommu: use atomic ops for irte index alloc
From: Peter Xu <peterx () redhat ! com>
Date: 2016-12-30 8:55:58
Message-ID: 1483088160-6714-7-git-send-email-peterx () redhat ! com
[Download RAW message or body]
To allow concurrent allocation of irte index. Meanwhile, move the IRTE
setup debug line into the alloc since vtd_setup_msi() might not be the
only one to dump this info in the future.
Signed-off-by: Peter Xu <peterx@redhat.com>
---
lib/x86/intel-iommu.c | 10 +++++++---
1 file changed, 7 insertions(+), 3 deletions(-)
diff --git a/lib/x86/intel-iommu.c b/lib/x86/intel-iommu.c
index 20b9240..a3ce678 100644
--- a/lib/x86/intel-iommu.c
+++ b/lib/x86/intel-iommu.c
@@ -13,6 +13,7 @@
#include "intel-iommu.h"
#include "libcflat.h"
#include "pci.h"
+#include "atomic.h"
/*
* VT-d in QEMU currently only support 39 bits address width, which is
@@ -239,9 +240,13 @@ void vtd_map_range(uint16_t sid, iova_t iova, phys_addr_t pa, size_t size)
static uint16_t vtd_intr_index_alloc(void)
{
- static int index_ctr = 0;
+ static volatile int index_ctr = 0;
+ int ctr;
+
assert(index_ctr < 65535);
- return index_ctr++;
+ ctr = atomic_inc_fetch(&index_ctr);
+ printf("INTR: alloc IRTE index %d\n", ctr);
+ return ctr;
}
static void vtd_setup_irte(struct pci_dev *dev, vtd_irte_t *irte,
@@ -296,7 +301,6 @@ bool vtd_setup_msi(struct pci_dev *dev, int vector, int dest_id)
assert(sizeof(vtd_msi_addr_t) == 8);
assert(sizeof(vtd_msi_data_t) == 4);
- printf("INTR: setup IRTE index %d\n", index);
vtd_setup_irte(dev, irte + index, vector, dest_id);
msi_addr.handle_15 = index >> 15 & 1;
--
2.7.4
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
[prev in list] [next in list] [prev in thread] [next in thread]
Configure |
About |
News |
Add a list |
Sponsored by KoreLogic