RE: [PATCH 14/20] IB/core: Add IB_DEVICE_OPA_MAD_SUPPORT device cap flag
> > On 1/12/2015 7:11 PM, ira.we...@intel.com wrote: > > Add a device capability flag to flag OPA MAD support on devices. > > You should put few words here telling what is OPA/OPA MADs and why > supporting them fits the IB core. > > See for example the IB core patch [1] that added signature verbs API is > states in > a manner which is both generic across vendors and across different Mellanox > HCAs. Also see [2] the IB core patch that added support for BMME API as > another example. I'm not quite sure what should be covered in this message which is different from the cover letter or the subsequent patches. Would this be more acceptable? IB/core: Add IB_DEVICE_OPA_MAD_SUPPORT device cap flag OPA MADs share a common header with IBTA MADs but with a different base version and an extended length. These "jumbo" MADs increase the performance of management traffic. Sharing a common header with IBTA MADs allows us to share most of the MAD processing code when dealing with OPA MADs in addition to supporting some IBTA MADs on OPA devices. Add a device capability flag to indicate OPA MAD support on the device. Ira > > > [1] 1b01d33 IB/core: Introduce signature verbs API [2] 00f7ec3 RDMA/core: > Add memory management extensions support -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
RE: [PATCH 14/20] IB/core: Add IB_DEVICE_OPA_MAD_SUPPORT device cap flag
> > On 1/12/2015 7:11 PM, ira.we...@intel.com wrote: > > --- a/include/rdma/ib_verbs.h > +++ b/include/rdma/ib_verbs.h > enum ib_signature_prot_cap { > IB_PROT_T10DIF_TYPE_1 = 1, > IB_PROT_T10DIF_TYPE_2 = 1 << 1, > @@ -210,6 +214,7 @@ struct ib_device_attr { > int sig_prot_cap; > int sig_guard_cap; > struct ib_odp_caps odp_caps; > + u64 device_cap_flags2; > > > Just make the existing kernel size device_cap_flags field a u64, note it's not > blankly copied to user space in uverbs as part of a chunk, so just go there > and > copy the lower 32 bits. > It is an easy change but before changing this I'd like to hear what Roland or others think. I prefer to define an additional field. Eventually we are likely to have flags which need to be communicated to user space. I think it will be cleaner at that time to report the entire new field rather than go through some sort of "upper 32 bit device capability" logic. Thoughts? Ira -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: [PATCH 14/20] IB/core: Add IB_DEVICE_OPA_MAD_SUPPORT device cap flag
On 1/12/2015 7:11 PM, ira.we...@intel.com wrote: Add a device capability flag to flag OPA MAD support on devices. You should put few words here telling what is OPA/OPA MADs and why supporting them fits the IB core. See for example the IB core patch [1] that added signature verbs API is states in a manner which is both generic across vendors and across different Mellanox HCAs. Also see [2] the IB core patch that added support for BMME API as another example. [1] 1b01d33 IB/core: Introduce signature verbs API [2] 00f7ec3 RDMA/core: Add memory management extensions support -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: [PATCH 14/20] IB/core: Add IB_DEVICE_OPA_MAD_SUPPORT device cap flag
On 1/12/2015 7:11 PM, ira.we...@intel.com wrote: --- a/include/rdma/ib_verbs.h +++ b/include/rdma/ib_verbs.h enum ib_signature_prot_cap { IB_PROT_T10DIF_TYPE_1 = 1, IB_PROT_T10DIF_TYPE_2 = 1 << 1, @@ -210,6 +214,7 @@ struct ib_device_attr { int sig_prot_cap; int sig_guard_cap; struct ib_odp_caps odp_caps; + u64 device_cap_flags2; Just make the existing kernel size device_cap_flags field a u64, note it's not blankly copied to user space in uverbs as part of a chunk, so just go there and copy the lower 32 bits. -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
[PATCH 14/20] IB/core: Add IB_DEVICE_OPA_MAD_SUPPORT device cap flag
From: Ira Weiny Add a device capability flag to flag OPA MAD support on devices. Signed-off-by: Ira Weiny --- Changes from RFC: Changed flag to indicate OPA MAD base version support Added device_cap_flags2 as device_cap_flags bits are limited include/rdma/ib_verbs.h |5 + 1 files changed, 5 insertions(+), 0 deletions(-) diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h index 5024f1f..83b39d5 100644 --- a/include/rdma/ib_verbs.h +++ b/include/rdma/ib_verbs.h @@ -128,6 +128,10 @@ enum ib_device_cap_flags { IB_DEVICE_ON_DEMAND_PAGING = (1<<31), }; +enum ib_device_cap_flags2 { + IB_DEVICE_OPA_MAD_SUPPORT = 1 +}; + enum ib_signature_prot_cap { IB_PROT_T10DIF_TYPE_1 = 1, IB_PROT_T10DIF_TYPE_2 = 1 << 1, @@ -210,6 +214,7 @@ struct ib_device_attr { int sig_prot_cap; int sig_guard_cap; struct ib_odp_caps odp_caps; + u64 device_cap_flags2; u32 mad_size; }; -- 1.7.1 -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html