[ofa-general] Re: [ewg] OFED Sep 10 meeting summary on OFED 1.3 development status

[Or Gerlitz]

Tziporet Koren wrote:
> OFED Sep 10 meeting summary on OFED 1.3 development status
> Meeting summary:
> 1. We reviewed OFED 1.3 features status:
> 	IPoIB: stateless offloads - 90% 

Tziporet, Michael

Based on the fact that OFED 1.3 is now based on 2.6.23-rc5, it simply 
does not make --any-- sense to have sixty "fix" patches from day one ( 
full list below). Zooming to to IPoIB (twelve patches!) I see plenty of 
non cooked materials:

The discussion/thread that followed the "IB/ipoib: S/G and HW checksum 
support" patch posted by Michael is not over yet. There was a direction 
offered by Michael which is pending to Roland's feedback.

As for the "ipoib - add LSO support" patch(es) posted by Eli, I did not 
see any response of Roland nor of anyone else (except for me asking some 
questions...). Eli/Michael - where does this stand? are you planning to 
push this for 2.6.24?

Also I understand that the patch set supports what you call "interrupt 
moderation", does this goes well with NAPI? if yes, how and why put in 
OFED something you never sent to review on the general list nor 
attempted to push into the kernel?

etc etc etc

> 	RDS - RDMA API - done 
> 	SDP: Keepalive - done; Asynch IO - done, Zero Copy - 80% 

have these patches ever sent to review on the general or ewg lists?

> 2. based on the status we decided to delay the feature freeze date to
> next week
> Alpha release is expected on Sep 19

I don't think we can start with sixty patches, sorry for not bringing 
this input before yesterday. I would be happy to hear what others here 
think.

Or.

> 0009-mlx4_add_wc.patch
> 0015_mlx4_set_cacheline_sz.patch
> 0018_mlx4_qp_per_mcg.patch
> 0020_mlx4_alloc_coherent.patch
> 0022-mlx4-mr-direct-mtt.patch
> 0023_mlx4_sg_stamp.patch
> 0024_mlx4_fmr.patch
> 0025_iw_cxgb3_writable-params.patch
> 0025-mlx4-sysfs-dev-info.patch
> 0026-mlx4-check-usr-sq-sz.patch
> 0027-mlx4-sqp-no-bitmap.patch
> 0028-cxgb3-fw-4.6.0
> cma_established1.patch
> cma_response_timeout.patch
> cma_tavor_quirk.patch
> cmd_tout.patch
> dma_map_sg.patch
> ehca_add_mutex_h.patch
> ipath-22-memcpy_cachebypass.patch
> ipoib_crash_wa.patch
> ipoib_dev_in_ipoib_neigh.patch
> ipoib_selector_updated.patch
> iwcm_ordird.patch
> mlx4_msix.patch
> mlx4_reset_msleep.patch
> mthca_catas_wqueue_namelen.patch
> mthca_msix.patch
> mthca_wrid_swap.patch
> mthca_x_qp_per_mcg.patch
> qos_0_mthca.patch
> scsi_transport_include_mutex.patch
> sdp_cq_param.patch
> sdp_post_credits.patch
> sean_cm_limit_mra_timeout.patch
> sean_local_sa_1_notifications.patch
> sean_local_sa_2_cache.patch
> sean_local_sa_3_disable.patch
> sean_local_sa_4_fix_hang.patch
> srp_1_recreate_at_reconnect.patch
> t_0010_ipoib_high_dma.patch
> t_0015_ipoib_sg.patch
> t_0020_core_csum.patch
> t_0030_mthca_checksum_offload.patch
> t_0040_mlx4_checksum_offload.patch
> t_0050_ipoib_checksum_offload.patch
> t_0060_ipoib_qp_init_attr.patch
> t_0080_mlx4_qp_max_msg.patch
> t_0090_core_lso.patch
> t_0100_mlx4_lso.patch
> t_0110_ipoib_lso.patch
> t_0120_ipoib_ethtool.patch
> t_0130_ipoib_lro.patch
> t_0140_core_modify_cq.patch
> t_0150_mlx4_modify_cq.patch
> t_0160_ipoib_modify_cq.patch
> t_0170_cq_coal.patch
> t_0180_ibcore_xrc.patch
> t_0181_ibcore_srq_create.patch
> t_0190_mlx4_xrc.patch
> t_0200_create_send_mad.patch