similar to: 6429600 freelist search algorithm misses some freelists in corner case

Author: kchow Repository: /hg/zfs-crypto/gate Revision: 6c5e7d7096ce47499922b13d2536fb929f6c843f Log message: 6286816 page_numtopp_nolock is inefficent 6288107 Missing comment terminator in i86pc/ml/locore.s 6290613 high page_get_cachelist failure rate on jurassic 6297005 performance issues with page_nextn() Files: update: usr/src/uts/common/os/mem_cage.c update: usr/src/uts/common/sys/cpuvar.h

Author: davemq Repository: /hg/zfs-crypto/gate Revision: 0542512405b483d0ffaf7d87d3fa6bf0dea49fe9 Log message: 6294795 The behavior of malloc is different between solaris 10 and nevada Files: update: usr/src/uts/i86pc/os/startup.c update: usr/src/uts/i86pc/vm/vm_machdep.c update: usr/src/uts/sun4/os/startup.c update: usr/src/uts/sun4/vm/vm_dep.h update: usr/src/uts/sun4u/vm/mach_vm_dep.c

Allow the iommu_unmap_fast to return newly freed page table pages and pass the freelist to queue_iova in the dma-iommu ops path. This is useful for iommu drivers (in this case the intel iommu driver) which need to wait for the ioTLB to be flushed before newly free/unmapped page table pages can be freed. This way we can still batch ioTLB free operations and handle the freelists. Signed-off-by:

Add a flush_iotlb_range to allow flushing of an iova range instead of a full flush in the dma-iommu path. Allow the iommu_unmap_fast to return newly freed page table pages and pass the freelist to queue_iova in the dma-iommu ops path. This patch is useful for iommu drivers (in this case the intel iommu driver) which need to wait for the ioTLB to be flushed before newly free/unmapped page table

On 2020-08-18 07:04, Tom Murphy wrote: > Add a flush_iotlb_range to allow flushing of an iova range instead of a > full flush in the dma-iommu path. > > Allow the iommu_unmap_fast to return newly freed page table pages and > pass the freelist to queue_iova in the dma-iommu ops path. > > This patch is useful for iommu drivers (in this case the intel iommu > driver) which

For supporting migration from VM, we need to have address_space on every page so zsmalloc shouldn't use page->mapping. So, this patch moves zs_meta from mapping to freelist. Signed-off-by: Minchan Kim <minchan at kernel.org> --- mm/zsmalloc.c | 22 +++++++++++----------- 1 file changed, 11 insertions(+), 11 deletions(-) diff --git a/mm/zsmalloc.c b/mm/zsmalloc.c index

Add a flush_iotlb_range to allow flushing of an iova range instead of a full flush in the dma-iommu path. Allow the iommu_unmap_fast to return newly freed page table pages and pass the freelist to queue_iova in the dma-iommu ops path. This patch is useful for iommu drivers (in this case the intel iommu driver) which need to wait for the ioTLB to be flushed before newly free/unmapped page table

Add a flush_iotlb_range to allow flushing of an iova range instead of a full flush in the dma-iommu path. Allow the iommu_unmap_fast to return newly freed page table pages and pass the freelist to queue_iova in the dma-iommu ops path. This patch is useful for iommu drivers (in this case the intel iommu driver) which need to wait for the ioTLB to be flushed before newly free/unmapped page table

Zsmalloc stores first free object's position into first_page->freelist in each zspage. If we change it with object index from first_page instead of location, we could squeeze it into page->mapping because the number of bit we need to store offset is at most 11bit. Signed-off-by: Minchan Kim <minchan at kernel.org> --- mm/zsmalloc.c | 159

Zsmalloc stores first free object's position into first_page->freelist in each zspage. If we change it with object index from first_page instead of location, we could squeeze it into page->mapping because the number of bit we need to store offset is at most 11bit. Signed-off-by: Minchan Kim <minchan at kernel.org> --- mm/zsmalloc.c | 158

From: Kent Overstreet <kmo at daterainc.com> Percpu frontend for allocating ids. With percpu allocation (that works), it's impossible to guarantee it will always be possible to allocate all nr_tags - typically, some will be stuck on a remote percpu freelist where the current job can't get to them. We do guarantee that it will always be possible to allocate at least (nr_tags / 2)

On Fri, 16 Aug 2013 23:09:06 +0000 "Nicholas A. Bellinger" <nab at linux-iscsi.org> wrote: > From: Kent Overstreet <kmo at daterainc.com> > > Percpu frontend for allocating ids. With percpu allocation (that works), > it's impossible to guarantee it will always be possible to allocate all > nr_tags - typically, some will be stuck on a remote percpu

On Fri, 16 Aug 2013 23:09:06 +0000 "Nicholas A. Bellinger" <nab at linux-iscsi.org> wrote: > From: Kent Overstreet <kmo at daterainc.com> > > Percpu frontend for allocating ids. With percpu allocation (that works), > it's impossible to guarantee it will always be possible to allocate all > nr_tags - typically, some will be stuck on a remote percpu

Author: ayznaga Repository: /hg/zfs-crypto/gate Revision: e3fd1c8e561df6513f95dba9c1ab58f4690d3354 Log message: PSARC 2005/738 SSG-Availability Feature Set 2 APR PSARC 2005/748 FMA for UltraSPARC DataPath 6229087 SPARC CPU/Mem DE should track DIMM serial numbers on Serengeti/Starcat 6281100 RFE: AVL-FS2: Support Anchored Page Retire for LW8, Serengeti & Starcat platforms 6281113 RFE: AVL-FS2:

On Tue, Aug 20, 2013 at 02:31:57PM -0700, Andrew Morton wrote: > On Fri, 16 Aug 2013 23:09:06 +0000 "Nicholas A. Bellinger" <nab at linux-iscsi.org> wrote: > > + /* > > + * Bitmap of cpus that (may) have tags on their percpu freelists: > > + * steal_tags() uses this to decide when to steal tags, and which cpus > > + * to try stealing from. > > +

Author: jchu Repository: /hg/zfs-crypto/gate Revision: 892c8f0a5eb063091eeecc146358651d83fc4d2f Log message: 6252398 Fire 2.0 Power Management support Files: update: usr/src/uts/sun4/io/px/pcie_pwr.c update: usr/src/uts/sun4/io/px/pcie_pwr.h update: usr/src/uts/sun4/io/px/px.c update: usr/src/uts/sun4/io/px/px_lib.h update: usr/src/uts/sun4/io/px/px_space.c update:

Author: schwartz Repository: /hg/zfs-crypto/gate Revision: b98644b3eadfc5ef90b11ebdabc347c8bc4fc386 Log message: 6350677 gcc and pcitool don''t get along anymore 6353954 pcitool for fire systems needs to deal better with the ranges property 6355025 pcitool probe crashes on chicago systems 6357523 sun4v pcitool phys access functions can''t be split across pages 6358836 pcitool nexus

Author: schwartz Repository: /hg/zfs-crypto/gate Revision: f5f3bfec5aa8a6e68f91ded399aa52be0424cd76 Log message: 6390254 fpc shouldn''t clutter sun4v unix with specialized hypervisor definitions 6390258 fpc should use hypervisor versioning APIs when available 6393704 pcitool dumps core with invalid options 6396857 fpc module causes boot-time panic on non-fire pcie hostbridge 6402441 sun4v

Author: cjj Repository: /hg/zfs-crypto/gate Revision: a58472277959c34c37827af84da191ceba990899 Log message: 6398779 Need to reduce address limits for devices downstream of a BCM 5708/5714/5715 bridge Files: create: usr/src/uts/sparc/pxb_bcm/Makefile update: usr/src/cmd/pcidr/etc/SUNW,EC_dr,ESC_dr_req,sysevent.conf update: usr/src/pkgdefs/SUNWckr/prototype_sparc update:

Full_Name: Nisai Wanakule Version: 1.7.1 OS: Solaris 2.9 Submission from: (NULL) (65.32.94.36) I tried to install RSPerl with --with-in-perl option and got some errors. R was built with --enable-R-shlib as indicated in the RSPerl instructions. Here is the list output during installation: nisai@lagrange:src 102$ R CMD INSTALL -c --configure-args='--with-in-perl' RSPerl_0.5-7.tar.gz *