On Altix, DMA may be reordered between a device and host memory. This reordering can happen in the NUMA interconnect, and it usually results in correct operation and improved performance. In some situations it may be necessary to explicitly synchronize DMA from the device. This patchset allows a memory region to be mapped with a "dmabarrier". Writes to the memory region will cause in-flight DMA to be flushed, providing a mechanism to order DMA from a device. There are 5 patches in this patchset: [1/5] dma: add dma_flags_set_attr() to dma interface [2/5] dma: redefine dma_flags_set_attr() for sn-ia64 [3/5] dma: document dma_flags_set_attr() [4/5] infiniband: add "dmabarrier" argument to ib_umem_get() [5/5] mthca: allow setting "dmabarrier" on user-allocated memory -- Arthur -
| Junio C Hamano | [ANNOUNCE] GIT 1.6.0 |
| Linus Torvalds | Re: [ANNOUNCE] mdb: Merkey's Linux Kernel Debugger 2.6.27-rc4 released |
| Bart Van Assche | Integration of SCST in the mainstream Linux kernel |
| Paul Menage | Re: [RFC][PATCH 6/7] Account for the number of tasks within container |
git: | |
| Nicolas Pitre | Re: pack operation is thrashing my server |
| Scott Chacon | Git Community Book |
| Greg KH | Re: [ANNOUNCE] pg - A patch porcelain for GIT |
| Lars Hjemli | [PATCH] git-merge: add option --no-ff |
| GVG GVG | ssh_exchange_identification: Connection closed by remote host |
| Nick Guenther | Re: Real men don't attack straw men |
| Travers Buda | Re: Important OpenBSD errata |
| Gregory Edigarov | How to re-build openssl with SHA1 support? |
| Al Boldi | [RFC] VM: I have a dream... |
| Dave Kleikamp | Re: [RFC] Heads up on sys_fallocate() |
| Jörn | Review status (Re: [PATCH] LogFS take three) |
| Chris Mason | [ANNOUNCE] Btrfs v0.12 released |
