On 04/27/2018 04:30 PM, Alan Tull wrote: > On Fri, Apr 27, 2018 at 1:26 PM, Florian Fainelli <f.faine...@gmail.com> > wrote: >> On 04/26/2018 06:26 PM, Moritz Fischer wrote: >>> From: Alan Tull <at...@kernel.org> >>> >>> Change fpga_mgr_register to not set or use drvdata. This supports >>> the case where a PCIe device has more than one manager. >>> >>> Add fpga_mgr_create/free functions. Change fpga_mgr_register and >>> fpga_mgr_unregister functions to take the mgr struct as their only >>> parameter. >>> >>> struct fpga_manager *fpga_mgr_create(struct device *dev, >>> const char *name, >>> const struct fpga_manager_ops *mops, >>> void *priv); >>> void fpga_mgr_free(struct fpga_manager *mgr); >>> int fpga_mgr_register(struct fpga_manager *mgr); >>> void fpga_mgr_unregister(struct fpga_manager *mgr); >>> >>> Update the drivers that call fpga_mgr_register with the new API. >> >> Apologies for chiming in so late, this commit does not make it clear >> that fpga_mgr_unregister() now also free the 'mgr' argument by calling >> fpga_mgr_free(), this is kind of detail, but an API should make that >> clear IMHO. > > If people follow the usage information, in > Documentation/fpga/fpga-mgr.txt, they'll do the right thing. But I > can add a patch that clarifies the description of fpga_mgr_unregister > in fpga-mgr.c that it "unregisters and frees" the manager.
Just mentioning that because not all APIs do this, take the network devices: there is an unregister_netdev() and a free_netdev(). Either way is fine with me as long as it is documented as such, I had to look at the API implementation to figure out that, no, all the drivers were not leaking their fpga_manager instance in their .remove() function :) -- Florian
