Re: [PATCH/RFC G-U-P experts] IB/umem: Modernize our get_user_pages() parameters

On Fri, 27 Jan 2012, Roland Dreier wrote: > > Sigh, what a mess ... it seems what we really want to do is know > > if userspace might trigger a COW because or not, and only do a > > preemptive COW in that case.  (We're not really concerned with > > userspace fork()ing and setting up a COW in the f

Re: [PATCH/RFC G-U-P experts] IB/umem: Modernize our get_user_pages() parameters

On Thu, 26 Jan 2012, Roland Dreier wrote: > > Thanks for the thoughtful answer... But I should have paid more attention to reading what you had written. In particular to: > > > This patch comes from me trying to do userspace RDMA on a memory > > > region exported from a character driver and mapp

Re: [PATCH] core/verb.c: fix kernel panic: always initialize struct ib_qp *qp->usecnt

On Fri, Jan 27, 2012 at 10:49 AM, Sven Breuner wrote: > thanks! Will you also take care of the Linux-stable (3.2.x) backport or > should I send it for inclusion in Linux-stable? I added the CC to stable in the commit so it should happen automatically. -- To unsubscribe from this list: send the li

Re: [PATCH] core/verb.c: fix kernel panic: always initialize struct ib_qp *qp->usecnt

Hi Roland, Roland Dreier wrote on 01/27/2012 06:20 PM: I've applied it and will send it to Linus soon. thanks! Will you also take care of the Linux-stable (3.2.x) backport or should I send it for inclusion in Linux-stable? Best regards, Sven -- To unsubscribe from this list: send the line

Re: [PATCH] rdma/ucma: Discard all events for new connections until accepted

Thanks for looking into this. applied... -- 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] IB/qib: unwind pcie change

Thanks, applied. -- 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 1/4] drivers/infiniband/hw/qib/qib_iba6120.c: convert GFP_KERNEL to GFP_ATOMIC

Thanks, applied. -- 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, RFC] nes: Add missing rcu_read_unlock() call

Good catch. I think it's closer to kernel idiom to jump to one unwind path, though: RDMA/nes: Add missing rcu_read_unlock() in nes_addr_resolve_neigh() Make sure all exit paths from this function unlock everything. Reported-by: Bart Van Assche Signed-off-by: Roland Dreier --- drivers/infinib

Re: [patch] infiniband: calling PTR_ERR() on wrong variable in create_file()

Thanks, applied. -- 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/RFC G-U-P experts] IB/umem: Modernize our get_user_pages() parameters

> Sigh, what a mess ... it seems what we really want to do is know > if userspace might trigger a COW because or not, and only do a > preemptive COW in that case.  (We're not really concerned with > userspace fork()ing and setting up a COW in the future, since that's > what we have MADV_DONTFORK fo

Re: [PATCH] core/verb.c: fix kernel panic: always initialize struct ib_qp *qp->usecnt

On Fri, Jan 27, 2012 at 8:23 AM, Sven Breuner wrote: > is there any progress on this? We have already reached Linux stable 3.2.2 > and IB still appears to be broken (at least for mthca) because of this. I've applied it and will send it to Linus soon. - R. -- To unsubscribe from this list: send

Re: [PATCH] core/verb.c: fix kernel panic: always initialize struct ib_qp *qp->usecnt

Hi, Roland Dreier wrote on 01/20/2012 07:43 PM: On Fri, Jan 20, 2012 at 10:40 AM, Roland Dreier wrote: On Fri, Jan 20, 2012 at 8:14 AM, Bernd Schubert wrote: I *guess* the qp allocated by pd->context->ops.create_qp() does not have qp->usecnt initialized (not does it know anything about it).