On Thu, Apr 01, 2021 at 02:47:04PM -0300, Jason Gunthorpe wrote: > On Thu, Mar 18, 2021 at 01:15:47PM +0200, Leon Romanovsky wrote: > > From: Maor Gottlieb <maorg@xxxxxxxxxx> > > > > MEMIC buffer, in addition to regular read and write operations, can > > support atomic operations from the host. > > > > Introduce and implement new UAPI to allocate address space for MEMIC > > operations such as atomic. This includes: <...> > It looks mostly fine otherwise, the error flows are a bit hard to read > though, when a new type is added this should also get re-organized so > we don't do stuff like: > > err_free: > /* In MEMIC error flow, dm will be freed internally */ > if (type != MLX5_IB_UAPI_DM_TYPE_MEMIC) > kfree(dm); I actually liked it, because the "re-organized" code was harder to read than this simple check. but ok, let's try again. Thanks