similar to: [PATCH 14/14] i386 / Introduce hypervisor lazy pinning hooks

By moving init_new_context and destroy_context inline into mmu_context.h, we can avoid extra functions calls, which are only needed for the unlikely case that the process context has an LDT to deal with. Now the code in ldt.c is called only when actually dealing with LDT creation or destruction. Careful analysis of alloc_ldt function showed that by using two const parameters, huge amounts of

By moving init_new_context and destroy_context inline into mmu_context.h, we can avoid extra functions calls, which are only needed for the unlikely case that the process context has an LDT to deal with. Now the code in ldt.c is called only when actually dealing with LDT creation or destruction. Careful analysis of alloc_ldt function showed that by using two const parameters, huge amounts of

I was also able to get the LDT switching functionality out of the critical path in switch_mm, which reduces the number of function calls, potential TLB misses and code size. Signed-off-by: Zachary Amsden <zach@vmware.com> Index: linux-2.6.14-zach-work/include/asm-i386/desc.h =================================================================== ---

I was also able to get the LDT switching functionality out of the critical path in switch_mm, which reduces the number of function calls, potential TLB misses and code size. Signed-off-by: Zachary Amsden <zach@vmware.com> Index: linux-2.6.14-zach-work/include/asm-i386/desc.h =================================================================== ---

* zach@vmware.com (zach@vmware.com) wrote: > Make the LDT a desc_struct pointer, since this is what it actually is. I like that plan. > There is code which relies on the fact that LDTs are allocated in page > chunks, and it is both cleaner and more convenient to keep the rather > poorly named "size" variable from the LDT in terms of LDT pages. I noticed it's replaced

* zach@vmware.com (zach@vmware.com) wrote: > Make the LDT a desc_struct pointer, since this is what it actually is. I like that plan. > There is code which relies on the fact that LDTs are allocated in page > chunks, and it is both cleaner and more convenient to keep the rather > poorly named "size" variable from the LDT in terms of LDT pages. I noticed it's replaced

Big cleanup of LDT code. This code has very little type checking and is not frequently used, so I audited the code, added type checking and size optimizations to generate smaller assembly code. I changed the ldt count to be in pages, and converted the char * ldt into a desc_struct. Signed-off-by: Zachary Amsden <zach@vmware.com> Index: linux-2.6.14-zach-work/arch/i386/kernel/ldt.c

Big cleanup of LDT code. This code has very little type checking and is not frequently used, so I audited the code, added type checking and size optimizations to generate smaller assembly code. I changed the ldt count to be in pages, and converted the char * ldt into a desc_struct. Signed-off-by: Zachary Amsden <zach@vmware.com> Index: linux-2.6.14-zach-work/arch/i386/kernel/ldt.c

Index: linux-2.6.13/arch/i386/kernel/ldt.c =================================================================== --- linux-2.6.13.orig/arch/i386/kernel/ldt.c 2005-08-08 15:27:26.000000000 -0700 +++ linux-2.6.13/arch/i386/kernel/ldt.c 2005-08-08 21:13:19.000000000 -0700 @@ -18,6 +18,7 @@ #include <asm/system.h> #include <asm/ldt.h> #include <asm/desc.h> +#include

Index: linux-2.6.13/arch/i386/kernel/ldt.c =================================================================== --- linux-2.6.13.orig/arch/i386/kernel/ldt.c 2005-08-08 15:27:26.000000000 -0700 +++ linux-2.6.13/arch/i386/kernel/ldt.c 2005-08-08 21:13:19.000000000 -0700 @@ -18,6 +18,7 @@ #include <asm/system.h> #include <asm/ldt.h> #include <asm/desc.h> +#include

Add Xen accessors for LDT updates. Todo: Bring back return values from write_ldt_entry. Signed-off-by: Zachary Amsden <zach@vmware.com> Index: linux-2.6.13/arch/i386/kernel/ldt.c =================================================================== --- linux-2.6.13.orig/arch/i386/kernel/ldt.c 2005-08-08 21:13:56.000000000 -0700 +++ linux-2.6.13/arch/i386/kernel/ldt.c 2005-08-08

Add Xen accessors for LDT updates. Todo: Bring back return values from write_ldt_entry. Signed-off-by: Zachary Amsden <zach@vmware.com> Index: linux-2.6.13/arch/i386/kernel/ldt.c =================================================================== --- linux-2.6.13.orig/arch/i386/kernel/ldt.c 2005-08-08 21:13:56.000000000 -0700 +++ linux-2.6.13/arch/i386/kernel/ldt.c 2005-08-08

Add hooks that the hypervisor can use to establish writable and non-writable pages for LDT pages. I made these parallel the page flags as defined in include/linux/page-flags.h, since the flag mechanism is very similar to the hypercall page flagging, and extended easily later to include PT, PD, PDP, GDT, etc. Patch-against: 2.6.13-rc5-mm1 Patch-keys: i386 ldt paravirt xen Signed-off-by: Zachary

Add hooks that the hypervisor can use to establish writable and non-writable pages for LDT pages. I made these parallel the page flags as defined in include/linux/page-flags.h, since the flag mechanism is very similar to the hypercall page flagging, and extended easily later to include PT, PD, PDP, GDT, etc. Patch-against: 2.6.13-rc5-mm1 Patch-keys: i386 ldt paravirt xen Signed-off-by: Zachary

The patched attached enables x86_64 xenlinux with "late pin, early unpin", which is already implemented for x86_32. Since we now only pin the root rather than any of the other levels, the overall performance became better especially with workloads that require heavy memory management operations. On 8-way x86_64 xenlinux (dom0) the kernel build was improved by about 10% (using make

I've got a fully updated Fedora Core 4 server crashing hard every week or two. I use Samba via smbmount and autofs to read & delete log files on 17 XP boxs and 6 NT4SP6 boxes as well as a couple other Windows files servers every 5 minutes. The first indication of a problem I get is smbmount stops working, then the server becomes unresponsive to the point where only a power slam will fix

These helpers are only for use with kernel threads, and I will tie them more into the kthread infrastructure going forward. Also move the prototypes to kthread.h - mmu_context.h was a little weird to start with as it otherwise contains very low-level MM bits. Signed-off-by: Christoph Hellwig <hch at lst.de> --- drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h | 1 +

vhost net module wants to do copy to/from user from a kernel thread, which needs use_mm (like what fs/aio has). Move that into mm/ and export to modules. Acked-by: Andrew Morton <akpm at linux-foundation.org> Signed-off-by: Michael S. Tsirkin <mst at redhat.com> --- fs/aio.c | 47 +---------------------------------- include/linux/mmu_context.h | 9 ++++++

vhost net module wants to do copy to/from user from a kernel thread, which needs use_mm (like what fs/aio has). Move that into mm/ and export to modules. Acked-by: Andrew Morton <akpm at linux-foundation.org> Signed-off-by: Michael S. Tsirkin <mst at redhat.com> --- fs/aio.c | 47 +---------------------------------- include/linux/mmu_context.h | 9 ++++++

vhost net module wants to do copy to/from user from a kernel thread, which needs use_mm (like what fs/aio has). Move that into mm/ and export to modules. Acked-by: Andrew Morton <akpm at linux-foundation.org> Acked-by: Andrea Arcangeli <aarcange at redhat.com> Signed-off-by: Michael S. Tsirkin <mst at redhat.com> --- fs/aio.c | 47