| Message ID | 1467795276-21725-19-git-send-email-peter.griffin@linaro.org |
|---|---|
| State | New |
| Headers | show |
On Wed, 06 Jul 2016, Peter Griffin wrote: > When rproc drivers are built-in the async firmware load done by rproc_add() > can fail due to the firmware not being present. Subsqeuent calls to > rproc_fw_boot() then fail, even though by this point firmware has been > successfully obtained. > > This patch changes the behaviour to re-execute rproc_fw_config_virtio() > in rproc_fw_boot() if it has previously failed, and we are sure it is > now available. > > Signed-off-by: Peter Griffin <peter.griffin@linaro.org> > --- > drivers/remoteproc/remoteproc_core.c | 25 +++++++++++++++++++------ > drivers/remoteproc/remoteproc_internal.h | 3 +++ > 2 files changed, 22 insertions(+), 6 deletions(-) > > diff --git a/drivers/remoteproc/remoteproc_core.c b/drivers/remoteproc/remoteproc_core.c > index db3958b..749f261 100644 > --- a/drivers/remoteproc/remoteproc_core.c > +++ b/drivers/remoteproc/remoteproc_core.c > @@ -801,8 +801,12 @@ static int rproc_fw_boot(struct rproc *rproc, const struct firmware *fw) > struct resource_table *table, *loaded_table; > int ret, tablesz; > > + /* > + * This can happen when built-in if initial async fw load fails. > + * However we now have firmware available so retry. > + */ > if (!rproc->table_ptr) > - return -ENOMEM; > + rproc_fw_config_virtio(fw, rproc); It's probably better to "hold that thought" on this patch, since we're going to make resource tables optional. I'm keen to tackle this issue, but need to work a few other things before doing so. Current task list: - Amend/supply new resource table from platform-specific drivers - Make resource tables optional - Obtain firmwares from the root file system - Lots more here including; co-proc DT support, co-proc console, etc [...] -- Lee Jones Linaro STMicroelectronics Landing Team Lead Linaro.org │ Open source software for ARM SoCs Follow Linaro: Facebook | Twitter | Blog
diff --git a/drivers/remoteproc/remoteproc_core.c b/drivers/remoteproc/remoteproc_core.c index db3958b..749f261 100644 --- a/drivers/remoteproc/remoteproc_core.c +++ b/drivers/remoteproc/remoteproc_core.c @@ -801,8 +801,12 @@ static int rproc_fw_boot(struct rproc *rproc, const struct firmware *fw) struct resource_table *table, *loaded_table; int ret, tablesz; + /* + * This can happen when built-in if initial async fw load fails. + * However we now have firmware available so retry. + */ if (!rproc->table_ptr) - return -ENOMEM; + rproc_fw_config_virtio(fw, rproc); ret = rproc_fw_sanity_check(rproc, fw); if (ret) @@ -895,9 +899,20 @@ clean_up: * to unregister the device. one other option is just to use kref here, * that might be cleaner). */ -static void rproc_fw_config_virtio(const struct firmware *fw, void *context) + +static void rproc_fw_config_virtio_cb(const struct firmware *fw, void *context) { struct rproc *rproc = context; + + rproc_fw_config_virtio(fw, rproc); + + release_firmware(fw); + /* allow rproc_del() contexts, if any, to proceed */ + complete_all(&rproc->firmware_loading_complete); +} + +static void rproc_fw_config_virtio(const struct firmware *fw, struct rproc *rproc) +{ struct resource_table *table; int ret, tablesz; @@ -934,9 +949,7 @@ static void rproc_fw_config_virtio(const struct firmware *fw, void *context) ret = rproc_handle_resources(rproc, tablesz, rproc_vdev_handler); out: - release_firmware(fw); - /* allow rproc_del() contexts, if any, to proceed */ - complete_all(&rproc->firmware_loading_complete); + return; } static int rproc_add_virtio_devices(struct rproc *rproc) @@ -956,7 +969,7 @@ static int rproc_add_virtio_devices(struct rproc *rproc) */ ret = request_firmware_nowait(THIS_MODULE, FW_ACTION_HOTPLUG, rproc->firmware, &rproc->dev, GFP_KERNEL, - rproc, rproc_fw_config_virtio); + rproc, rproc_fw_config_virtio_cb); if (ret < 0) { dev_err(&rproc->dev, "request_firmware_nowait err: %d\n", ret); complete_all(&rproc->firmware_loading_complete); diff --git a/drivers/remoteproc/remoteproc_internal.h b/drivers/remoteproc/remoteproc_internal.h index 57e1de5..39916c1 100644 --- a/drivers/remoteproc/remoteproc_internal.h +++ b/drivers/remoteproc/remoteproc_internal.h @@ -49,6 +49,9 @@ struct rproc_fw_ops { void rproc_release(struct kref *kref); irqreturn_t rproc_vq_interrupt(struct rproc *rproc, int vq_id); int rproc_boot_nowait(struct rproc *rproc); +static void rproc_fw_config_virtio_cb(const struct firmware *fw, void *context); +static void rproc_fw_config_virtio(const struct firmware *fw, struct rproc *rproc); + /* from remoteproc_virtio.c */ int rproc_add_virtio_dev(struct rproc_vdev *rvdev, int id);
When rproc drivers are built-in the async firmware load done by rproc_add() can fail due to the firmware not being present. Subsqeuent calls to rproc_fw_boot() then fail, even though by this point firmware has been successfully obtained. This patch changes the behaviour to re-execute rproc_fw_config_virtio() in rproc_fw_boot() if it has previously failed, and we are sure it is now available. Signed-off-by: Peter Griffin <peter.griffin@linaro.org> --- drivers/remoteproc/remoteproc_core.c | 25 +++++++++++++++++++------ drivers/remoteproc/remoteproc_internal.h | 3 +++ 2 files changed, 22 insertions(+), 6 deletions(-) -- 1.9.1