* [dpdk-dev] [PATCH] eal/linux: fix multi-process cannot work
@ 2017-03-16 16:28 Jianfeng Tan
2017-03-29 15:19 ` Sergio Gonzalez Monroy
0 siblings, 1 reply; 3+ messages in thread
From: Jianfeng Tan @ 2017-03-16 16:28 UTC (permalink / raw)
To: dev; +Cc: sergio.gonzalez.monroy, benjamin.walker, Jianfeng Tan, stable
When binding with vfio-pci, secondary process cannot be started with
an error message:
cannot find TAILQ entry for PCI device.
It's due to: struct rte_pci_addr is padded with 1 byte for alignment
by compiler. Then below comparison in commit 2f4adfad0a69
("vfio: add multiprocess support") will fail if the last byte is not
initialized.
memcmp(&vfio_res->pci_addr, &dev->addr, sizeof(dev->addr)
And commit cdc242f260e7 ("eal/linux: support running as unprivileged user")
just triggers this bug by using a stack un-initialized variable.
The fix is to use rte_eal_compare_pci_addr() for pci addr comparison.
Fixes: 2f4adfad0a69 ("vfio: add multiprocess support")
Fixes: cdc242f260e7 ("eal/linux: support running as unprivileged user")
CC: stable@dpdk.org
Reported-by: Rutkowski, Pawel <pawelx.rutkowski@intel.com>
Signed-off-by: Jianfeng Tan <jianfeng.tan@intel.com>
---
lib/librte_eal/linuxapp/eal/eal_pci_vfio.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/lib/librte_eal/linuxapp/eal/eal_pci_vfio.c b/lib/librte_eal/linuxapp/eal/eal_pci_vfio.c
index 5f478c5..7d8b9fb 100644
--- a/lib/librte_eal/linuxapp/eal/eal_pci_vfio.c
+++ b/lib/librte_eal/linuxapp/eal/eal_pci_vfio.c
@@ -355,7 +355,8 @@ pci_vfio_map_resource(struct rte_pci_device *dev)
} else {
/* if we're in a secondary process, just find our tailq entry */
TAILQ_FOREACH(vfio_res, vfio_res_list, next) {
- if (memcmp(&vfio_res->pci_addr, &dev->addr, sizeof(dev->addr)))
+ if (rte_eal_compare_pci_addr(&vfio_res->pci_addr,
+ &dev->addr))
continue;
break;
}
--
2.7.4
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [dpdk-dev] [PATCH] eal/linux: fix multi-process cannot work
2017-03-16 16:28 [dpdk-dev] [PATCH] eal/linux: fix multi-process cannot work Jianfeng Tan
@ 2017-03-29 15:19 ` Sergio Gonzalez Monroy
2017-04-04 9:58 ` Thomas Monjalon
0 siblings, 1 reply; 3+ messages in thread
From: Sergio Gonzalez Monroy @ 2017-03-29 15:19 UTC (permalink / raw)
To: Jianfeng Tan, dev; +Cc: benjamin.walker, stable
On 16/03/2017 16:28, Jianfeng Tan wrote:
> When binding with vfio-pci, secondary process cannot be started with
> an error message:
>
> cannot find TAILQ entry for PCI device.
>
> It's due to: struct rte_pci_addr is padded with 1 byte for alignment
> by compiler. Then below comparison in commit 2f4adfad0a69
> ("vfio: add multiprocess support") will fail if the last byte is not
> initialized.
>
> memcmp(&vfio_res->pci_addr, &dev->addr, sizeof(dev->addr)
>
> And commit cdc242f260e7 ("eal/linux: support running as unprivileged user")
> just triggers this bug by using a stack un-initialized variable.
>
> The fix is to use rte_eal_compare_pci_addr() for pci addr comparison.
>
> Fixes: 2f4adfad0a69 ("vfio: add multiprocess support")
> Fixes: cdc242f260e7 ("eal/linux: support running as unprivileged user")
> CC: stable@dpdk.org
>
> Reported-by: Rutkowski, Pawel <pawelx.rutkowski@intel.com>
> Signed-off-by: Jianfeng Tan <jianfeng.tan@intel.com>
> ---
> lib/librte_eal/linuxapp/eal/eal_pci_vfio.c | 3 ++-
> 1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/lib/librte_eal/linuxapp/eal/eal_pci_vfio.c b/lib/librte_eal/linuxapp/eal/eal_pci_vfio.c
> index 5f478c5..7d8b9fb 100644
> --- a/lib/librte_eal/linuxapp/eal/eal_pci_vfio.c
> +++ b/lib/librte_eal/linuxapp/eal/eal_pci_vfio.c
> @@ -355,7 +355,8 @@ pci_vfio_map_resource(struct rte_pci_device *dev)
> } else {
> /* if we're in a secondary process, just find our tailq entry */
> TAILQ_FOREACH(vfio_res, vfio_res_list, next) {
> - if (memcmp(&vfio_res->pci_addr, &dev->addr, sizeof(dev->addr)))
> + if (rte_eal_compare_pci_addr(&vfio_res->pci_addr,
> + &dev->addr))
> continue;
> break;
> }
Acked-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [dpdk-dev] [PATCH] eal/linux: fix multi-process cannot work
2017-03-29 15:19 ` Sergio Gonzalez Monroy
@ 2017-04-04 9:58 ` Thomas Monjalon
0 siblings, 0 replies; 3+ messages in thread
From: Thomas Monjalon @ 2017-04-04 9:58 UTC (permalink / raw)
To: Jianfeng Tan; +Cc: dev, Sergio Gonzalez Monroy, benjamin.walker, stable
2017-03-29 16:19, Sergio Gonzalez Monroy:
> On 16/03/2017 16:28, Jianfeng Tan wrote:
> > When binding with vfio-pci, secondary process cannot be started with
> > an error message:
> >
> > cannot find TAILQ entry for PCI device.
> >
> > It's due to: struct rte_pci_addr is padded with 1 byte for alignment
> > by compiler. Then below comparison in commit 2f4adfad0a69
> > ("vfio: add multiprocess support") will fail if the last byte is not
> > initialized.
> >
> > memcmp(&vfio_res->pci_addr, &dev->addr, sizeof(dev->addr)
> >
> > And commit cdc242f260e7 ("eal/linux: support running as unprivileged user")
> > just triggers this bug by using a stack un-initialized variable.
> >
> > The fix is to use rte_eal_compare_pci_addr() for pci addr comparison.
> >
> > Fixes: 2f4adfad0a69 ("vfio: add multiprocess support")
> > Fixes: cdc242f260e7 ("eal/linux: support running as unprivileged user")
> > CC: stable@dpdk.org
> >
> > Reported-by: Rutkowski, Pawel <pawelx.rutkowski@intel.com>
> > Signed-off-by: Jianfeng Tan <jianfeng.tan@intel.com>
>
> Acked-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
Applied, thanks
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2017-04-04 9:58 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2017-03-16 16:28 [dpdk-dev] [PATCH] eal/linux: fix multi-process cannot work Jianfeng Tan
2017-03-29 15:19 ` Sergio Gonzalez Monroy
2017-04-04 9:58 ` 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).