[dpdk-dev,1/2] eal_interrupts: support external rxq interrupt handlers
Checks
Commit Message
Prior to this patch only UIO/VFIO interrupt handlers types were supported.
This patch adds support for the external interrupt handler type, allowing
external drivers to set their own fds with specific interrupt handlers.
Signed-off-by: Shahaf Shuler <shahafs@mellanox.com>
---
lib/librte_eal/linuxapp/eal/eal_interrupts.c | 33 ++++++++++++++--------
.../linuxapp/eal/include/exec-env/rte_interrupts.h | 9 ++++++
2 files changed, 30 insertions(+), 12 deletions(-)
Comments
2017-02-08 15:57, Shahaf Shuler:
> /**
> + * It deletes registered eventfds.
> + *
> + * @param intr_handle
> + * Pointer to the interrupt handle.
> + */
> +void
> +rte_intr_free_epoll_fd(struct rte_intr_handle *intr_handle);
This patch looks OK except a miss in the EAL map files for this new function.
The title should be "eal/linux: support external Rx interrupt"
I think this patch should target next-net.
Please Ferruh, take care of this patchset. Thanks
Thursday, March 9, 2017 8:10 PM, Thomas Monjalon:
> 2017-02-08 15:57, Shahaf Shuler:
> > /**
> > + * It deletes registered eventfds.
> > + *
> > + * @param intr_handle
> > + * Pointer to the interrupt handle.
> > + */
> > +void
> > +rte_intr_free_epoll_fd(struct rte_intr_handle *intr_handle);
>
> This patch looks OK except a miss in the EAL map files for this new function.
> The title should be "eal/linux: support external Rx interrupt"
>
> I think this patch should target next-net.
> Please Ferruh, take care of this patchset. Thanks
Ferruh, any more comments before submitting a v2?
On 3/12/2017 8:30 AM, Shahaf Shuler wrote:
> Thursday, March 9, 2017 8:10 PM, Thomas Monjalon:
>> 2017-02-08 15:57, Shahaf Shuler:
>>> /**
>>> + * It deletes registered eventfds.
>>> + *
>>> + * @param intr_handle
>>> + * Pointer to the interrupt handle.
>>> + */
>>> +void
>>> +rte_intr_free_epoll_fd(struct rte_intr_handle *intr_handle);
>>
>> This patch looks OK except a miss in the EAL map files for this new function.
>> The title should be "eal/linux: support external Rx interrupt"
>>
>> I think this patch should target next-net.
>> Please Ferruh, take care of this patchset. Thanks
>
> Ferruh, any more comments before submitting a v2?
Hi Shahaf,
No extra comments, only second patch does not cleanly apply to next-net,
please rebase it to before sending.
Thanks,
ferruh
@@ -937,6 +937,8 @@ static __attribute__((noreturn)) void *
bytes_read = sizeof(buf.vfio_intr_count);
break;
#endif
+ case RTE_INTR_HANDLE_EXT:
+ return;
default:
bytes_read = 1;
RTE_LOG(INFO, EAL, "unexpected intr type\n");
@@ -1165,6 +1167,24 @@ static __attribute__((noreturn)) void *
return rc;
}
+void
+rte_intr_free_epoll_fd(struct rte_intr_handle *intr_handle)
+{
+ uint32_t i;
+ struct rte_epoll_event *rev;
+
+ for (i = 0; i < intr_handle->nb_efd; i++) {
+ rev = &intr_handle->elist[i];
+ if (rev->status == RTE_EPOLL_INVALID)
+ continue;
+ if (rte_epoll_ctl(rev->epfd, EPOLL_CTL_DEL, rev->fd, rev)) {
+ /* force free if the entry valid */
+ eal_epoll_data_safe_free(rev);
+ rev->status = RTE_EPOLL_INVALID;
+ }
+ }
+}
+
int
rte_intr_efd_enable(struct rte_intr_handle *intr_handle, uint32_t nb_efd)
{
@@ -1200,19 +1220,8 @@ static __attribute__((noreturn)) void *
rte_intr_efd_disable(struct rte_intr_handle *intr_handle)
{
uint32_t i;
- struct rte_epoll_event *rev;
-
- for (i = 0; i < intr_handle->nb_efd; i++) {
- rev = &intr_handle->elist[i];
- if (rev->status == RTE_EPOLL_INVALID)
- continue;
- if (rte_epoll_ctl(rev->epfd, EPOLL_CTL_DEL, rev->fd, rev)) {
- /* force free if the entry valid */
- eal_epoll_data_safe_free(rev);
- rev->status = RTE_EPOLL_INVALID;
- }
- }
+ rte_intr_free_epoll_fd(intr_handle);
if (intr_handle->max_intr > intr_handle->nb_efd) {
for (i = 0; i < intr_handle->nb_efd; i++)
close(intr_handle->efds[i]);
@@ -171,6 +171,15 @@ struct rte_intr_handle {
int epfd, int op, unsigned int vec, void *data);
/**
+ * It deletes registered eventfds.
+ *
+ * @param intr_handle
+ * Pointer to the interrupt handle.
+ */
+void
+rte_intr_free_epoll_fd(struct rte_intr_handle *intr_handle);
+
+/**
* It enables the packet I/O interrupt event if it's necessary.
* It creates event fd for each interrupt vector when MSIX is used,
* otherwise it multiplexes a single event fd.