On 11/25/07, Lukas Hejtmanek <[EMAIL PROTECTED]> wrote: > I've already resolved this issue. For some reason, kernel drivers (2.6.18) > works correctly under Xen Dom0. I was confused about OFED 1.2 which works but > it contained patch to work under Xen. After applying similar patch to the OFED > 1.3a, it works now as well.
Can you please clarify what is the patch in OFED 1.2 which once deployed to OFED 1.3 makes it work for you? Looking on the 1.2 sources under ofa_kernel-1.2/kernel_patches/fixes I see three patches that contain the word coherent, but neither of them seemed to me what you are referring to, am I wrong? The patches are: mthca_3_alloc_consistent.patch mthca_4_dma_align_reserved_mtts.patch mthca_5_roland_fix_icm_coherent.patch Or. change-log for mthca_3_alloc_consistent.patch Fix non-cache-coherent CPUs with memfree HCAs. We allocate the MTT table with alloc_pages() and then do pci_map_sg(), so we must call pci_dma_sync_sg after the CPU writes to the MTT table (this works since device never writes the MTTs on memfree). For MPTs, both the device and CPU might write there, so we must allocate dma coherent memory for these. Signed-off-by: Michael S. Tsirkin <[EMAIL PROTECTED]> change-log for mthca_4_dma_align_reserved_mtts.patch [PATCHv3 5 of 5] IB/mthca: reserved MTTs and memory alignment issues This fixes several issues related to reserved MTTs and memory alignment. 1. MTTs are allocated in non-cache-coherent memory, so we must give reserved MTTs their own cache line, to prevent both device and CPU from writing into the same cache line at the same time. 2. reserved_mtts field has different meaning in Tavor and Arbel, so we are wasting mtt entries on memfree. fix the Arbel case to match Tavor semantics. Signed-off-by: Michael S. Tsirkin <[EMAIL PROTECTED]> change-log for mthca_5_roland_fix_icm_coherent.patch commit 1c7c25e1a36ea2aa053373dfb363fd646677edac Author: Roland Dreier <[EMAIL PROTECTED]> Date: Wed Feb 14 08:45:17 2007 -0800 IB/mthca: Fix allocation of ICM chunks in coherent memory The change to allow allocating ICM chunks from coherent memory did not increment the count of sg entries properly, so a chunk that required more than allocation would not be mapped properly by the HCA. Fix this by adding the missing increment of chunk->nsg. Signed-off-by: Roland Dreier <[EMAIL PROTECTED]> _______________________________________________ general mailing list [email protected] http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
