DPDK patches and discussions
 help / color / mirror / Atom feed
* [dpdk-dev] [PATCH] pci: fix closing an unopened file descriptor
@ 2013-09-04  7:34 Tetsuya Mukawa
  2013-09-04  7:38 ` Tetsuya.Mukawa
  2013-09-12 15:28 ` Thomas Monjalon
  0 siblings, 2 replies; 3+ messages in thread
From: Tetsuya Mukawa @ 2013-09-04  7:34 UTC (permalink / raw)
  To: dev

If CONFIG_RTE_EAL_UNBIND_PORTS is set and virtio-net is used, an unopened
file descriptor will be illegally closed in the finalized phase of EAL.
The fix adds a correct initial value to the file descriptor, and check it
before closing it.

Signed-off-by: Tetsuya Mukawa <mukawa@igel.co.jp>
---
 lib/librte_eal/linuxapp/eal/eal_pci.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/lib/librte_eal/linuxapp/eal/eal_pci.c b/lib/librte_eal/linuxapp/eal/eal_pci.c
index c793148..7c04ba3 100644
--- a/lib/librte_eal/linuxapp/eal/eal_pci.c
+++ b/lib/librte_eal/linuxapp/eal/eal_pci.c
@@ -519,6 +519,8 @@ pci_scan_one(const char *dirname, uint16_t domain, uint8_t bus,
 	dev->addr.devid = devid;
 	dev->addr.function = function;
 
+	dev->intr_handle.fd = -1;
+
 	/* get vendor id */
 	rte_snprintf(filename, sizeof(filename), "%s/vendor", dirname);
 	if (eal_parse_sysfs_value(filename, &tmp) < 0) {
@@ -718,7 +720,8 @@ pci_exit_process(struct rte_pci_device *dev)
 		RTE_LOG(ERR, EAL, "Error with munmap\n");
 		return -1;
 	}
-	if (close(dev->intr_handle.fd) == -1){
+	if ((dev->intr_handle.fd != -1) &&
+			(close(dev->intr_handle.fd) == -1)) {
 		RTE_LOG(ERR, EAL, "Error closing interrupt handle\n");
 		return -1;
 	}
-- 
1.8.3.1

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [dpdk-dev] [PATCH] pci: fix closing an unopened file descriptor
  2013-09-04  7:34 [dpdk-dev] [PATCH] pci: fix closing an unopened file descriptor Tetsuya Mukawa
@ 2013-09-04  7:38 ` Tetsuya.Mukawa
  2013-09-12 15:28 ` Thomas Monjalon
  1 sibling, 0 replies; 3+ messages in thread
From: Tetsuya.Mukawa @ 2013-09-04  7:38 UTC (permalink / raw)
  To: dev, thomas.monjalon

Hi Thomas,

I've sent the patch to fix an uninitialized variable access.
Could you please check it?
I will send ver.2 patch if I need.

Thanks,
Tetsuya Mukawa

(2013/09/04 16:34), Tetsuya Mukawa wrote:
> If CONFIG_RTE_EAL_UNBIND_PORTS is set and virtio-net is used, an unopened
> file descriptor will be illegally closed in the finalized phase of EAL.
> The fix adds a correct initial value to the file descriptor, and check it
> before closing it.
>
> Signed-off-by: Tetsuya Mukawa <mukawa@igel.co.jp>
> ---
>  lib/librte_eal/linuxapp/eal/eal_pci.c | 5 ++++-
>  1 file changed, 4 insertions(+), 1 deletion(-)
>
> diff --git a/lib/librte_eal/linuxapp/eal/eal_pci.c b/lib/librte_eal/linuxapp/eal/eal_pci.c
> index c793148..7c04ba3 100644
> --- a/lib/librte_eal/linuxapp/eal/eal_pci.c
> +++ b/lib/librte_eal/linuxapp/eal/eal_pci.c
> @@ -519,6 +519,8 @@ pci_scan_one(const char *dirname, uint16_t domain, uint8_t bus,
>  	dev->addr.devid = devid;
>  	dev->addr.function = function;
>  
> +	dev->intr_handle.fd = -1;
> +
>  	/* get vendor id */
>  	rte_snprintf(filename, sizeof(filename), "%s/vendor", dirname);
>  	if (eal_parse_sysfs_value(filename, &tmp) < 0) {
> @@ -718,7 +720,8 @@ pci_exit_process(struct rte_pci_device *dev)
>  		RTE_LOG(ERR, EAL, "Error with munmap\n");
>  		return -1;
>  	}
> -	if (close(dev->intr_handle.fd) == -1){
> +	if ((dev->intr_handle.fd != -1) &&
> +			(close(dev->intr_handle.fd) == -1)) {
>  		RTE_LOG(ERR, EAL, "Error closing interrupt handle\n");
>  		return -1;
>  	}

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [dpdk-dev] [PATCH] pci: fix closing an unopened file descriptor
  2013-09-04  7:34 [dpdk-dev] [PATCH] pci: fix closing an unopened file descriptor Tetsuya Mukawa
  2013-09-04  7:38 ` Tetsuya.Mukawa
@ 2013-09-12 15:28 ` Thomas Monjalon
  1 sibling, 0 replies; 3+ messages in thread
From: Thomas Monjalon @ 2013-09-12 15:28 UTC (permalink / raw)
  To: Tetsuya Mukawa; +Cc: dev

04/09/2013 09:34, Tetsuya Mukawa :
> If CONFIG_RTE_EAL_UNBIND_PORTS is set and virtio-net is used, an unopened
> file descriptor will be illegally closed in the finalized phase of EAL.
> The fix adds a correct initial value to the file descriptor, and check it
> before closing it.
> 
> Signed-off-by: Tetsuya Mukawa <mukawa@igel.co.jp>

Thanks for the patch.

I have reproduced the issue only once.
It seems that fd 0 is open most of the time.

But the patch seems OK so I applied it with light modifications:

---

pci: fix closing an unopened file descriptor
    
If CONFIG_RTE_EAL_UNBIND_PORTS is set and a non Intel PMD is used, an unopened
file descriptor will be illegally closed in the finalized phase of EAL.
The fix adds a correct initial value to the file descriptor, and check it
before closing it.
    
Signed-off-by: Tetsuya Mukawa <mukawa@igel.co.jp>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>

--- a/lib/librte_eal/linuxapp/eal/eal_pci.c
+++ b/lib/librte_eal/linuxapp/eal/eal_pci.c
@@ -514,6 +514,7 @@ pci_scan_one(const char *dirname, uint16_t domain, uint8_t 
bus,
 	}
 
 	memset(dev, 0, sizeof(*dev));
+	dev->intr_handle.fd = -1;
 	dev->addr.domain = domain;
 	dev->addr.bus = bus;
 	dev->addr.devid = devid;
@@ -718,7 +719,7 @@ pci_exit_process(struct rte_pci_device *dev)
 		RTE_LOG(ERR, EAL, "Error with munmap\n");
 		return -1;
 	}
-	if (close(dev->intr_handle.fd) == -1){
+	if ((dev->intr_handle.fd != -1) && (close(dev->intr_handle.fd) == -1)) {
 		RTE_LOG(ERR, EAL, "Error closing interrupt handle\n");
 		return -1;
 	}

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2013-09-12 15:28 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2013-09-04  7:34 [dpdk-dev] [PATCH] pci: fix closing an unopened file descriptor Tetsuya Mukawa
2013-09-04  7:38 ` Tetsuya.Mukawa
2013-09-12 15:28 ` Thomas Monjalon

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).