On Thu, 2010-08-19 at 23:10 -0500, Sonny Rao wrote:
> the eHCA driver registers a MR for all of kernel memory, but makes the
> assumption that valid memory exists at KERNELBASE. This assumption
> may not be true in the case of a relocatable kernel, so use KERNELBASE
> + PHYSICAL_START to get the true beginning of usable kernel memory.
>
> This patch depends on the earlier patch which exports the necessary
> symbol for PHYSICAL_START in a relocatable kernel.
I'm going to send patch 1/2 to Linus today or tomorrow, so this patch
can then be picked up by the infiniband maintainers.
Cheers,
Ben.
> cc: Joachim Fenkes <fen...@de.ibm.com>
> cc: Christoph Raisch <rai...@de.ibm.com>
> cc: Hoan-Ham Hguyen <hngu...@de.ibm.com>
> Signed-off-by: Sonny Rao <sonny...@us.ibm.com>
>
>
> Index: linux-2.6/drivers/infiniband/hw/ehca/ehca_mrmw.c
> ===================================================================
> --- linux-2.6.orig/drivers/infiniband/hw/ehca/ehca_mrmw.c 2010-08-09
> 22:16:57.688652613 -0500
> +++ linux-2.6/drivers/infiniband/hw/ehca/ehca_mrmw.c 2010-08-19
> 22:53:03.451507146 -0500
> @@ -171,7 +171,7 @@
> }
>
> ret = ehca_reg_maxmr(shca, e_maxmr,
> - (void *)ehca_map_vaddr((void *)KERNELBASE),
> + (void *)ehca_map_vaddr((void *)(KERNELBASE
> + PHYSICAL_START)),
> mr_access_flags, e_pd,
> &e_maxmr->ib.ib_mr.lkey,
> &e_maxmr->ib.ib_mr.rkey);
> @@ -1636,7 +1636,7 @@
>
> /* register internal max-MR on HCA */
> size_maxmr = ehca_mr_len;
> - iova_start = (u64 *)ehca_map_vaddr((void *)KERNELBASE);
> + iova_start = (u64 *)ehca_map_vaddr((void *)(KERNELBASE +
> PHYSICAL_START));
> ib_pbuf.addr = 0;
> ib_pbuf.size = size_maxmr;
> num_kpages = NUM_CHUNKS(((u64)iova_start % PAGE_SIZE) + size_maxmr,
> @@ -2209,7 +2209,7 @@
> {
> /* a MR is treated as max-MR only if it fits following: */
> if ((size == ehca_mr_len) &&
> - (iova_start == (void *)ehca_map_vaddr((void *)KERNELBASE))) {
> + (iova_start == (void *)ehca_map_vaddr((void *)(KERNELBASE +
> PHYSICAL_START)))) {
> ehca_gen_dbg("this is a max-MR");
> return 1;
> } else
> _______________________________________________
> Linuxppc-dev mailing list
> linuxppc-...@lists.ozlabs.org
> https://lists.ozlabs.org/listinfo/linuxppc-dev
--
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
