[PATCH 00/21] X86_64, UV: Update kernel for SGI UV4 support

2016-04-29 Thread Mike Travis
This patch set primarily updates the Linux kernel to support the next generation SGI Ultraviolet system, UV4. This architecture change is a larger incremental change than previous UV updates because of a major change to the addressing scheme. Previous UV architectures used a fixed address width

[PATCH 11/21] X86_64, UV: Move scir info to the per cpu info struct

2016-04-29 Thread Mike Travis
Change the references to the SCIR fields to the new per cpu info structs. Signed-off-by: Mike Travis <tra...@sgi.com> Reviewed-by: Dimitri Sivanich <sivan...@sgi.com> Tested-by: John Estabrook <estabr...@sgi.com> Tested-by: Gary Kroening <g...@sgi.com> Tested-by: Nathan

[PATCH 13/21] X86_64, UV: Allocate common per node hub info structs on local node

2016-04-29 Thread Mike Travis
when a group of cpus on a node are all interrupted for a common task. Signed-off-by: Mike Travis <tra...@sgi.com> Reviewed-by: Dimitri Sivanich <sivan...@sgi.com> Reviewed-by: Andrew Banman <aban...@sgi.com> Tested-by: John Estabrook <estabr...@sgi.com> Tested-by: Gary Kroeni

[PATCH 11/21] X86_64, UV: Move scir info to the per cpu info struct

2016-04-29 Thread Mike Travis
Change the references to the SCIR fields to the new per cpu info structs. Signed-off-by: Mike Travis Reviewed-by: Dimitri Sivanich Tested-by: John Estabrook Tested-by: Gary Kroening Tested-by: Nathan Zimmer --- arch/x86/include/asm/uv/uv_hub.h | 17 ++--- arch/x86/kernel

[PATCH 13/21] X86_64, UV: Allocate common per node hub info structs on local node

2016-04-29 Thread Mike Travis
when a group of cpus on a node are all interrupted for a common task. Signed-off-by: Mike Travis Reviewed-by: Dimitri Sivanich Reviewed-by: Andrew Banman Tested-by: John Estabrook Tested-by: Gary Kroening Tested-by: Nathan Zimmer --- arch/x86/include/asm/uv/uv_hub.h | 36

[PATCH 15/21] X86_64, UV: Add UV4 addressing discovery function

2016-04-29 Thread Mike Travis
. Signed-off-by: Mike Travis <tra...@sgi.com> Reviewed-by: Dimitri Sivanich <sivan...@sgi.com> Tested-by: John Estabrook <estabr...@sgi.com> Tested-by: Gary Kroening <g...@sgi.com> Tested-by: Nathan Zimmer <nzim...@sgi.com> --- arch/x86/ke

[PATCH 20/21] X86_64, UV: Remove Obsolete GRU MMR address translation

2016-04-29 Thread Mike Travis
From: Dimitri Sivanich <sivan...@sgi.com> Use no-op messages in place of cross-partition interrupts when nacking a put message in the GRU. This allows us to remove MMR's as a destination from the GRU driver. Signed-off-by: Dimitri Sivanich <sivan...@sgi.com> Signed-off-by: Mike

[PATCH 20/21] X86_64, UV: Remove Obsolete GRU MMR address translation

2016-04-29 Thread Mike Travis
From: Dimitri Sivanich Use no-op messages in place of cross-partition interrupts when nacking a put message in the GRU. This allows us to remove MMR's as a destination from the GRU driver. Signed-off-by: Dimitri Sivanich Signed-off-by: Mike Travis Tested-by: John Estabrook Tested-by: Gary

[PATCH 15/21] X86_64, UV: Add UV4 addressing discovery function

2016-04-29 Thread Mike Travis
. Signed-off-by: Mike Travis Reviewed-by: Dimitri Sivanich Tested-by: John Estabrook Tested-by: Gary Kroening Tested-by: Nathan Zimmer --- arch/x86/kernel/apic/x2apic_uv_x.c | 85 +++-- 1 file changed, 72 insertions(+), 13 deletions(-) --- linux.orig/arch/x86

[PATCH 08/21] X86_64, UV: Clean up redunduncies after merge of UV4 MMR definitions

2016-04-29 Thread Mike Travis
Clean up any redundancies caused by new UV4 MMR definitions superseding any previously definitions local to functions. Signed-off-by: Mike Travis <tra...@sgi.com> Reviewed-by: Dimitri Sivanich <sivan...@sgi.com> Reviewed-by: Andrew Banman <aban...@sgi.com> Tested-by: John

[PATCH 08/21] X86_64, UV: Clean up redunduncies after merge of UV4 MMR definitions

2016-04-29 Thread Mike Travis
Clean up any redundancies caused by new UV4 MMR definitions superseding any previously definitions local to functions. Signed-off-by: Mike Travis Reviewed-by: Dimitri Sivanich Reviewed-by: Andrew Banman Tested-by: John Estabrook Tested-by: Gary Kroening Tested-by: Nathan Zimmer --- arch

[PATCH 09/21] X86_64, UV: Update MMIOH setup function to work for both UV3 and UV4

2016-04-29 Thread Mike Travis
Since UV3 and UV4 MMIOH regions are setup the same, we can use a common function to setup both. Signed-off-by: Mike Travis <tra...@sgi.com> Reviewed-by: Dimitri Sivanich <sivan...@sgi.com> Tested-by: John Estabrook <estabr...@sgi.com> Tested-by: Gary Kroening <g...@sgi.c

[PATCH 04/21] X86_64, UV: Add UV MMR Illegal Access Function

2016-04-29 Thread Mike Travis
the problem. It allows functions common to some UV arches to use common defines that might not be fully defined for all arches, as long as they do not reference them on the unsupported arches. Signed-off-by: Mike Travis <tra...@sgi.com> Reviewed-by: Dimitri Sivanich <sivan...@sgi.co

[PATCH 09/21] X86_64, UV: Update MMIOH setup function to work for both UV3 and UV4

2016-04-29 Thread Mike Travis
Since UV3 and UV4 MMIOH regions are setup the same, we can use a common function to setup both. Signed-off-by: Mike Travis Reviewed-by: Dimitri Sivanich Tested-by: John Estabrook Tested-by: Gary Kroening Tested-by: Nathan Zimmer --- arch/x86/kernel/apic/x2apic_uv_x.c |3 ++- 1 file

[PATCH 04/21] X86_64, UV: Add UV MMR Illegal Access Function

2016-04-29 Thread Mike Travis
the problem. It allows functions common to some UV arches to use common defines that might not be fully defined for all arches, as long as they do not reference them on the unsupported arches. Signed-off-by: Mike Travis Reviewed-by: Dimitri Sivanich Tested-by: John Estabrook Tested-by: Gary

[PATCH 16/21] X86_64, UV: Add obtaining GAM Range Table from UV BIOS

2016-04-29 Thread Mike Travis
V BIOS via the EFI UVsystab table. Support for older EFI UVsystab tables is maintained. Signed-off-by: Mike Travis <tra...@sgi.com> Tested-by: Dimitri Sivanich <sivan...@sgi.com> Tested-by: John Estabrook <estabr...@sgi.com> Tested-by: Gary Kroening <g...@sgi.com> Tested-

[PATCH 16/21] X86_64, UV: Add obtaining GAM Range Table from UV BIOS

2016-04-29 Thread Mike Travis
V BIOS via the EFI UVsystab table. Support for older EFI UVsystab tables is maintained. Signed-off-by: Mike Travis Tested-by: Dimitri Sivanich Tested-by: John Estabrook Tested-by: Gary Kroening Tested-by: Nathan Zimmer --- arch/x86/include/asm/uv/bios

[PATCH 02/21] X86_64, UV: Add UV Architecture Defines

2016-04-29 Thread Mike Travis
Add defines to control which UV architectures are supported, and modify the 'if (is_uvX_*)' functions to return constant 0 for those not supported. This will help optimize code paths when support for specific UV arches is removed. Signed-off-by: Mike Travis <tra...@sgi.com> Reviewed-by: D

[PATCH 14/21] X86_64, UV: Fold blade info into per node hub info structs

2016-04-29 Thread Mike Travis
in the comments and whitespace to clean things up a bit. Signed-off-by: Mike Travis <tra...@sgi.com> Tested-by: Dimitri Sivanich <sivan...@sgi.com> Reviewed-by: Andrew Banman <aban...@sgi.com> Tested-by: John Estabrook <estabr...@sgi.com> Tested-by: Gary Kroening <g...@sgi.com>

[PATCH 02/21] X86_64, UV: Add UV Architecture Defines

2016-04-29 Thread Mike Travis
Add defines to control which UV architectures are supported, and modify the 'if (is_uvX_*)' functions to return constant 0 for those not supported. This will help optimize code paths when support for specific UV arches is removed. Signed-off-by: Mike Travis Reviewed-by: Dimitri Sivanich Tested

[PATCH 14/21] X86_64, UV: Fold blade info into per node hub info structs

2016-04-29 Thread Mike Travis
in the comments and whitespace to clean things up a bit. Signed-off-by: Mike Travis Tested-by: Dimitri Sivanich Reviewed-by: Andrew Banman Tested-by: John Estabrook Tested-by: Gary Kroening Tested-by: Nathan Zimmer --- arch/x86/include/asm/uv/uv_hub.h | 71 +--- arch/x86/kernel/apic

[PATCH 21/21] X86_64, UV: Fix incorrect nodes and pnodes for cpuless and memoryless nodes

2016-04-29 Thread Mike Travis
From: Dimitri Sivanich <sivan...@sgi.com> This patch fixes the problem of incorrect nodes and pnodes being returned when referring to nodes that either have no cpus (AKA "headless") or no memory. Signed-off-by: Dimitri Sivanich <sivan...@sgi.com> Signed-off-by: Mike

[PATCH 17/21] X86_64, UV: Support UV4 socket address changes

2016-04-29 Thread Mike Travis
With the UV4 system architecture addressing changes, BIOS now provides this information via an EFI system table. This is the initial decoding of that system table. It also collects the sizing information for later allocation of dynamic conversion tables. Signed-off-by: Mike Travis <

[PATCH 21/21] X86_64, UV: Fix incorrect nodes and pnodes for cpuless and memoryless nodes

2016-04-29 Thread Mike Travis
From: Dimitri Sivanich This patch fixes the problem of incorrect nodes and pnodes being returned when referring to nodes that either have no cpus (AKA "headless") or no memory. Signed-off-by: Dimitri Sivanich Signed-off-by: Mike Travis Tested-by: John Estabrook Tested-by: Gar

[PATCH 17/21] X86_64, UV: Support UV4 socket address changes

2016-04-29 Thread Mike Travis
With the UV4 system architecture addressing changes, BIOS now provides this information via an EFI system table. This is the initial decoding of that system table. It also collects the sizing information for later allocation of dynamic conversion tables. Signed-off-by: Mike Travis Reviewed

[PATCH 19/21] X86_64, UV: Update physical address conversions for UV4

2016-04-29 Thread Mike Travis
to measure execution time. Signed-off-by: Mike Travis <tra...@sgi.com> Tested-by: Dimitri Sivanich <sivan...@sgi.com> Tested-by: John Estabrook <estabr...@sgi.com> Tested-by: Gary Kroening <g...@sgi.com> Tested-by: Nathan Zimmer <nzim...@sgi.com> --- arch/x86/in

[PATCH 05/21] X86_64, UV: Prep for UV4 MMR updates

2016-04-29 Thread Mike Travis
overlay addresses and NMI definitions to allow for flexibility in newer UV architecture types. Signed-off-by: Mike Travis <tra...@sgi.com> Reviewed-by: Dimitri Sivanich <sivan...@sgi.com> Tested-by: John Estabrook <estabr...@sgi.com> Tested-by: Gary Kroening <g...@sgi.com>

[PATCH 07/21] X86_64, UV: Disable Obsolete APIC ID fixup code used only on UV1

2016-04-29 Thread Mike Travis
macro. It will also remove other dead code that is conditioned by "if (is_uv1_hub()) ..." for those systems that do not need nor want that extra code. Signed-off-by: Mike Travis <tra...@sgi.com> Reviewed-by: Dimitri Sivanich <sivan...@sgi.com> Reviewed-by: Russ Anderson <

[PATCH 19/21] X86_64, UV: Update physical address conversions for UV4

2016-04-29 Thread Mike Travis
to measure execution time. Signed-off-by: Mike Travis Tested-by: Dimitri Sivanich Tested-by: John Estabrook Tested-by: Gary Kroening Tested-by: Nathan Zimmer --- arch/x86/include/asm/uv/uv_hub.h | 131 - arch/x86/kernel/apic/x2apic_uv_x.c | 96

[PATCH 05/21] X86_64, UV: Prep for UV4 MMR updates

2016-04-29 Thread Mike Travis
overlay addresses and NMI definitions to allow for flexibility in newer UV architecture types. Signed-off-by: Mike Travis Reviewed-by: Dimitri Sivanich Tested-by: John Estabrook Tested-by: Gary Kroening Tested-by: Nathan Zimmer --- arch/x86/include/asm/uv/uv_hub.h |5 arch/x86/kernel

[PATCH 07/21] X86_64, UV: Disable Obsolete APIC ID fixup code used only on UV1

2016-04-29 Thread Mike Travis
macro. It will also remove other dead code that is conditioned by "if (is_uv1_hub()) ..." for those systems that do not need nor want that extra code. Signed-off-by: Mike Travis Reviewed-by: Dimitri Sivanich Reviewed-by: Russ Anderson Tested-by: Nathan Zimmer --- arch/x86/Kc

[PATCH 18/21] X86_64, UV: Build GAM reference tables

2016-04-29 Thread Mike Travis
An aspect of the UV4 system architecture changes involve changing the way sockets, nodes, and pnodes are translated between one another. Decode the information from the BIOS provided EFI system table to build the needed conversion tables. Signed-off-by: Mike Travis <tra...@sgi.com> Re

[PATCH 18/21] X86_64, UV: Build GAM reference tables

2016-04-29 Thread Mike Travis
An aspect of the UV4 system architecture changes involve changing the way sockets, nodes, and pnodes are translated between one another. Decode the information from the BIOS provided EFI system table to build the needed conversion tables. Signed-off-by: Mike Travis Reviewed-by: Dimitri Sivanich

[PATCH 03/21] X86_64, UV: Add UV4 Specific Defines

2016-04-29 Thread Mike Travis
Add UV4 specific defines to determine if current system type is a UV4 system. Signed-off-by: Mike Travis <tra...@sgi.com> Reviewed-by: Dimitri Sivanich <sivan...@sgi.com> Tested-by: John Estabrook <estabr...@sgi.com> Tested-by: Gary Kroening <g...@sgi.com> Tested

[PATCH 03/21] X86_64, UV: Add UV4 Specific Defines

2016-04-29 Thread Mike Travis
Add UV4 specific defines to determine if current system type is a UV4 system. Signed-off-by: Mike Travis Reviewed-by: Dimitri Sivanich Tested-by: John Estabrook Tested-by: Gary Kroening Tested-by: Nathan Zimmer --- arch/x86/kernel/apic/x2apic_uv_x.c | 12 +--- 1 file changed, 9

[PATCH 07/21] X86_64, UV: Enable/Disable UV1 only workaround code

2016-04-28 Thread Mike Travis
is was due to only having 8 APIC ID bits thus a workaround was added to add and remove extra bits to the APIC ID. It also required a specialized UV1 only apic driver. Signed-off-by: Mike Travis <tra...@sgi.com> Tested-by: Nathan Zimmer <nzim...@sgi.com> --- arch/x86/Kconfig

[PATCH 07/21] X86_64, UV: Enable/Disable UV1 only workaround code

2016-04-28 Thread Mike Travis
is was due to only having 8 APIC ID bits thus a workaround was added to add and remove extra bits to the APIC ID. It also required a specialized UV1 only apic driver. Signed-off-by: Mike Travis Tested-by: Nathan Zimmer --- arch/x86/Kconfig | 14 ++ arch/x86/incl

[PATCH 05/21] X86_64, UV: Prep for UV4 MMR updates

2016-04-28 Thread Mike Travis
overlay addresses and NMI definitions to allow for flexibility in newer UV architecture types. Signed-off-by: Mike Travis <tra...@sgi.com> Reviewed-by: Dimitri Sivanich <sivan...@sgi.com> Tested-by: John Estabrook <estabr...@sgi.com> Tested-by: Gary Kroening <g...@sgi.com>

[PATCH 05/21] X86_64, UV: Prep for UV4 MMR updates

2016-04-28 Thread Mike Travis
overlay addresses and NMI definitions to allow for flexibility in newer UV architecture types. Signed-off-by: Mike Travis Reviewed-by: Dimitri Sivanich Tested-by: John Estabrook Tested-by: Gary Kroening Tested-by: Nathan Zimmer --- arch/x86/include/asm/uv/uv_hub.h |5 arch/x86/kernel

[PATCH 09/21] X86_64, UV: Update MMIOH setup function to work for both UV3 and UV4

2016-04-28 Thread Mike Travis
Since UV3 and UV4 MMIOH regions are setup the same, we can use a common function to setup both. Signed-off-by: Mike Travis <tra...@sgi.com> Reviewed-by: Dimitri Sivanich <sivan...@sgi.com> Tested-by: John Estabrook <estabr...@sgi.com> Tested-by: Gary Kroening <g...@sgi.c

[PATCH 09/21] X86_64, UV: Update MMIOH setup function to work for both UV3 and UV4

2016-04-28 Thread Mike Travis
Since UV3 and UV4 MMIOH regions are setup the same, we can use a common function to setup both. Signed-off-by: Mike Travis Reviewed-by: Dimitri Sivanich Tested-by: John Estabrook Tested-by: Gary Kroening Tested-by: Nathan Zimmer --- arch/x86/kernel/apic/x2apic_uv_x.c |3 ++- 1 file

[PATCH 00/21] X86_64, UV: Update kernel for SGI UV4 support

2016-04-28 Thread Mike Travis
This patch set primarily updates the Linux kernel to support the next generation SGI Ultraviolet system, UV4. This architecture change is a larger incremental change than previous UV updates because of a major change to the addressing scheme. Previous UV architectures used a fixed address width

[PATCH 00/21] X86_64, UV: Update kernel for SGI UV4 support

2016-04-28 Thread Mike Travis
This patch set primarily updates the Linux kernel to support the next generation SGI Ultraviolet system, UV4. This architecture change is a larger incremental change than previous UV updates because of a major change to the addressing scheme. Previous UV architectures used a fixed address width

[PATCH 03/21] X86_64, UV: Add UV4 Specific Defines

2016-04-28 Thread Mike Travis
Add UV4 specific defines to determine if current system type is a UV4 system. Signed-off-by: Mike Travis <tra...@sgi.com> Reviewed-by: Dimitri Sivanich <sivan...@sgi.com> Tested-by: John Estabrook <estabr...@sgi.com> Tested-by: Gary Kroening <g...@sgi.com> Tested

[PATCH 03/21] X86_64, UV: Add UV4 Specific Defines

2016-04-28 Thread Mike Travis
Add UV4 specific defines to determine if current system type is a UV4 system. Signed-off-by: Mike Travis Reviewed-by: Dimitri Sivanich Tested-by: John Estabrook Tested-by: Gary Kroening Tested-by: Nathan Zimmer --- arch/x86/kernel/apic/x2apic_uv_x.c | 12 +--- 1 file changed, 9

[PATCH 08/21] X86_64, UV: Clean up redunduncies after merge of UV4 MMR definitions

2016-04-28 Thread Mike Travis
Clean up any redundancies caused by new UV4 MMR definitions superseding any previously definitions local to functions. Signed-off-by: Mike Travis <tra...@sgi.com> Reviewed-by: Dimitri Sivanich <sivan...@sgi.com> Reviewed-by: Andrew Banman <aban...@sgi.com> Tested-by: John

[PATCH 12/21] X86_64, UV: Move blade local processor ID to the per cpu info struct

2016-04-28 Thread Mike Travis
definition. This allows calling source code to be built for both pre-UV4 kernels as well as post-UV4 kernels. Signed-off-by: Mike Travis <tra...@sgi.com> Reviewed-by: Dimitri Sivanich <sivan...@sgi.com> Tested-by: John Estabrook <estabr...@sgi.com> Tested-by: Gary Kroening <

[PATCH 13/21] X86_64, UV: Allocate common per node hub info structs on local node

2016-04-28 Thread Mike Travis
when a group of cpus on a node are all interrupted for a common task. Signed-off-by: Mike Travis <tra...@sgi.com> Reviewed-by: Dimitri Sivanich <sivan...@sgi.com> Reviewed-by: Andrew Banman <aban...@sgi.com> Tested-by: John Estabrook <estabr...@sgi.com> Tested-by: Gary Kroeni

[PATCH 04/21] X86_64, UV: Add UV MMR Illegal Access Function

2016-04-28 Thread Mike Travis
the problem. It allows functions common to some UV arches to use common defines that might not be fully defined for all arches, as long as they do not reference them on the unsupported arches. Signed-off-by: Mike Travis <tra...@sgi.com> Reviewed-by: Dimitri Sivanich <sivan...@sgi.co

[PATCH 16/21] X86_64, UV: Add obtaining GAM Range Table from UV BIOS

2016-04-28 Thread Mike Travis
V BIOS via the EFI UVsystab table. Support for older EFI UVsystab tables is maintained. Signed-off-by: Mike Travis <tra...@sgi.com> Tested-by: Dimitri Sivanich <sivan...@sgi.com> Tested-by: John Estabrook <estabr...@sgi.com> Tested-by: Gary Kroening <g...@sgi.com> Tested-

[PATCH 08/21] X86_64, UV: Clean up redunduncies after merge of UV4 MMR definitions

2016-04-28 Thread Mike Travis
Clean up any redundancies caused by new UV4 MMR definitions superseding any previously definitions local to functions. Signed-off-by: Mike Travis Reviewed-by: Dimitri Sivanich Reviewed-by: Andrew Banman Tested-by: John Estabrook Tested-by: Gary Kroening Tested-by: Nathan Zimmer --- arch

[PATCH 12/21] X86_64, UV: Move blade local processor ID to the per cpu info struct

2016-04-28 Thread Mike Travis
definition. This allows calling source code to be built for both pre-UV4 kernels as well as post-UV4 kernels. Signed-off-by: Mike Travis Reviewed-by: Dimitri Sivanich Tested-by: John Estabrook Tested-by: Gary Kroening Tested-by: Nathan Zimmer --- arch/x86/include/asm/uv/uv_hub.h | 10

[PATCH 13/21] X86_64, UV: Allocate common per node hub info structs on local node

2016-04-28 Thread Mike Travis
when a group of cpus on a node are all interrupted for a common task. Signed-off-by: Mike Travis Reviewed-by: Dimitri Sivanich Reviewed-by: Andrew Banman Tested-by: John Estabrook Tested-by: Gary Kroening Tested-by: Nathan Zimmer --- arch/x86/include/asm/uv/uv_hub.h | 36

[PATCH 04/21] X86_64, UV: Add UV MMR Illegal Access Function

2016-04-28 Thread Mike Travis
the problem. It allows functions common to some UV arches to use common defines that might not be fully defined for all arches, as long as they do not reference them on the unsupported arches. Signed-off-by: Mike Travis Reviewed-by: Dimitri Sivanich Tested-by: John Estabrook Tested-by: Gary

[PATCH 16/21] X86_64, UV: Add obtaining GAM Range Table from UV BIOS

2016-04-28 Thread Mike Travis
V BIOS via the EFI UVsystab table. Support for older EFI UVsystab tables is maintained. Signed-off-by: Mike Travis Tested-by: Dimitri Sivanich Tested-by: John Estabrook Tested-by: Gary Kroening Tested-by: Nathan Zimmer --- arch/x86/include/asm/uv/bios

[PATCH 18/21] X86_64, UV: Build GAM reference tables

2016-04-28 Thread Mike Travis
An aspect of the UV4 system architecture changes involve changing the way sockets, nodes, and pnodes are translated between one another. Decode the information from the BIOS provided EFI system table to build the needed conversion tables. Signed-off-by: Mike Travis <tra...@sgi.com> Re

[PATCH 14/21] X86_64, UV: Fold blade info into per node hub info structs

2016-04-28 Thread Mike Travis
in the comments and whitespace to clean things up a bit. Signed-off-by: Mike Travis <tra...@sgi.com> Tested-by: Dimitri Sivanich <sivan...@sgi.com> Reviewed-by: Andrew Banman <aban...@sgi.com> Tested-by: John Estabrook <estabr...@sgi.com> Tested-by: Gary Kroening <g...@sgi.com>

[PATCH 18/21] X86_64, UV: Build GAM reference tables

2016-04-28 Thread Mike Travis
An aspect of the UV4 system architecture changes involve changing the way sockets, nodes, and pnodes are translated between one another. Decode the information from the BIOS provided EFI system table to build the needed conversion tables. Signed-off-by: Mike Travis Reviewed-by: Dimitri Sivanich

[PATCH 14/21] X86_64, UV: Fold blade info into per node hub info structs

2016-04-28 Thread Mike Travis
in the comments and whitespace to clean things up a bit. Signed-off-by: Mike Travis Tested-by: Dimitri Sivanich Reviewed-by: Andrew Banman Tested-by: John Estabrook Tested-by: Gary Kroening Tested-by: Nathan Zimmer --- arch/x86/include/asm/uv/uv_hub.h | 71 +--- arch/x86/kernel/apic

[PATCH 19/21] X86_64, UV: Update physical address conversions for UV4

2016-04-28 Thread Mike Travis
to measure execution time. Signed-off-by: Mike Travis <tra...@sgi.com> Tested-by: Dimitri Sivanich <sivan...@sgi.com> Tested-by: John Estabrook <estabr...@sgi.com> Tested-by: Gary Kroening <g...@sgi.com> Tested-by: Nathan Zimmer <nzim...@sgi.com> --- arch/x86/in

[PATCH 21/21] X86_64, UV: Fix incorrect nodes and pnodes for cpuless and memoryless nodes

2016-04-28 Thread Mike Travis
From: Dimitri Sivanich <sivan...@sgi.com> This patch fixes the problem of incorrect nodes and pnodes being returned when referring to nodes that either have no cpus (AKA "headless") or no memory. Signed-off-by: Dimitri Sivanich <sivan...@sgi.com> Signed-off-by: Mike

[PATCH 20/21] X86_64, UV: Remove Obsolete GRU MMR address translation

2016-04-28 Thread Mike Travis
From: Dimitri Sivanich <sivan...@sgi.com> Use no-op messages in place of cross-partition interrupts when nacking a put message in the GRU. This allows us to remove MMR's as a destination from the GRU driver. Signed-off-by: Dimitri Sivanich <sivan...@sgi.com> Signed-off-by: Mike

[PATCH 17/21] X86_64, UV: Support UV4 socket address changes

2016-04-28 Thread Mike Travis
With the UV4 system architecture addressing changes, BIOS now provides this information via an EFI system table. This is the initial decoding of that system table. It also collects the sizing information for later allocation of dynamic conversion tables. Signed-off-by: Mike Travis <

[PATCH 15/21] X86_64, UV: Add UV4 addressing discovery function

2016-04-28 Thread Mike Travis
. Signed-off-by: Mike Travis <tra...@sgi.com> Reviewed-by: Dimitri Sivanich <sivan...@sgi.com> Tested-by: John Estabrook <estabr...@sgi.com> Tested-by: Gary Kroening <g...@sgi.com> Tested-by: Nathan Zimmer <nzim...@sgi.com> --- arch/x86/ke

[PATCH 01/21] X86_64, UV: Add Initial UV4 definitions

2016-04-28 Thread Mike Travis
Add preliminary UV4 defines. Signed-off-by: Mike Travis <tra...@sgi.com> Tested-by: John Estabrook <estabr...@sgi.com> Tested-by: Gary Kroening <g...@sgi.com> Tested-by: Nathan Zimmer <nzim...@sgi.com> --- arch/x86/include/asm/uv/uv_hub.h | 64

[PATCH 19/21] X86_64, UV: Update physical address conversions for UV4

2016-04-28 Thread Mike Travis
to measure execution time. Signed-off-by: Mike Travis Tested-by: Dimitri Sivanich Tested-by: John Estabrook Tested-by: Gary Kroening Tested-by: Nathan Zimmer --- arch/x86/include/asm/uv/uv_hub.h | 131 - arch/x86/kernel/apic/x2apic_uv_x.c | 96

[PATCH 21/21] X86_64, UV: Fix incorrect nodes and pnodes for cpuless and memoryless nodes

2016-04-28 Thread Mike Travis
From: Dimitri Sivanich This patch fixes the problem of incorrect nodes and pnodes being returned when referring to nodes that either have no cpus (AKA "headless") or no memory. Signed-off-by: Dimitri Sivanich Signed-off-by: Mike Travis Tested-by: John Estabrook Tested-by: Gar

[PATCH 20/21] X86_64, UV: Remove Obsolete GRU MMR address translation

2016-04-28 Thread Mike Travis
From: Dimitri Sivanich Use no-op messages in place of cross-partition interrupts when nacking a put message in the GRU. This allows us to remove MMR's as a destination from the GRU driver. Signed-off-by: Dimitri Sivanich Signed-off-by: Mike Travis Tested-by: John Estabrook Tested-by: Gary

[PATCH 17/21] X86_64, UV: Support UV4 socket address changes

2016-04-28 Thread Mike Travis
With the UV4 system architecture addressing changes, BIOS now provides this information via an EFI system table. This is the initial decoding of that system table. It also collects the sizing information for later allocation of dynamic conversion tables. Signed-off-by: Mike Travis Reviewed

[PATCH 15/21] X86_64, UV: Add UV4 addressing discovery function

2016-04-28 Thread Mike Travis
. Signed-off-by: Mike Travis Reviewed-by: Dimitri Sivanich Tested-by: John Estabrook Tested-by: Gary Kroening Tested-by: Nathan Zimmer --- arch/x86/kernel/apic/x2apic_uv_x.c | 85 +++-- 1 file changed, 72 insertions(+), 13 deletions(-) --- linux.orig/arch/x86

[PATCH 01/21] X86_64, UV: Add Initial UV4 definitions

2016-04-28 Thread Mike Travis
Add preliminary UV4 defines. Signed-off-by: Mike Travis Tested-by: John Estabrook Tested-by: Gary Kroening Tested-by: Nathan Zimmer --- arch/x86/include/asm/uv/uv_hub.h | 64 +++--- arch/x86/include/asm/uv/uv_mmrs.h | 22 + 2 files changed, 62

[PATCH 10/21] X86_64, UV: Create per cpu info structs to replace per hub info structs

2016-04-28 Thread Mike Travis
The major portion of the hub info is common to all cpus on that hub. This is step one of moving the per cpu hub info to a per node hub info struct. This patch creates the small per cpu info struct that will contain only information specific to each CPU. Signed-off-by: Mike Travis <tra...@sgi.

[PATCH 02/21] X86_64, UV: Add UV Architecture Defines

2016-04-28 Thread Mike Travis
Add defines to control which UV architectures are supported, and modify the 'if (is_uvX_*)' functions to return constant 0 for those not supported. This will help optimize code paths when support for specific UV arches is removed. Signed-off-by: Mike Travis <tra...@sgi.com> Reviewed-by: D

[PATCH 02/21] X86_64, UV: Add UV Architecture Defines

2016-04-28 Thread Mike Travis
Add defines to control which UV architectures are supported, and modify the 'if (is_uvX_*)' functions to return constant 0 for those not supported. This will help optimize code paths when support for specific UV arches is removed. Signed-off-by: Mike Travis Reviewed-by: Dimitri Sivanich Tested

[PATCH 10/21] X86_64, UV: Create per cpu info structs to replace per hub info structs

2016-04-28 Thread Mike Travis
The major portion of the hub info is common to all cpus on that hub. This is step one of moving the per cpu hub info to a per node hub info struct. This patch creates the small per cpu info struct that will contain only information specific to each CPU. Signed-off-by: Mike Travis Reviewed

[PATCH 11/21] X86_64, UV: Move scir info to the per cpu info struct

2016-04-28 Thread Mike Travis
Change the references to the SCIR fields to the new per cpu info structs. Signed-off-by: Mike Travis <tra...@sgi.com> Reviewed-by: Dimitri Sivanich <sivan...@sgi.com> Tested-by: John Estabrook <estabr...@sgi.com> Tested-by: Gary Kroening <g...@sgi.com> Tested-by: Nathan

[PATCH 11/21] X86_64, UV: Move scir info to the per cpu info struct

2016-04-28 Thread Mike Travis
Change the references to the SCIR fields to the new per cpu info structs. Signed-off-by: Mike Travis Reviewed-by: Dimitri Sivanich Tested-by: John Estabrook Tested-by: Gary Kroening Tested-by: Nathan Zimmer --- arch/x86/include/asm/uv/uv_hub.h | 17 ++--- arch/x86/kernel

Re: [PATCH 07/21] X86_64, UV: Remove Support for UV1

2016-04-28 Thread Mike Travis
"checkpatches" script also had some new "concerns". I was able to fix all that were possible (or practical). The patches follow... Thanks! Mike On 4/28/2016 1:48 AM, Ingo Molnar wrote: > > * Mike Travis <tra...@sgi.com> wrote: > >> The UV1 architectu

Re: [PATCH 07/21] X86_64, UV: Remove Support for UV1

2016-04-28 Thread Mike Travis
"checkpatches" script also had some new "concerns". I was able to fix all that were possible (or practical). The patches follow... Thanks! Mike On 4/28/2016 1:48 AM, Ingo Molnar wrote: > > * Mike Travis wrote: > >> The UV1 architecture is no longer being su

[PATCH 01/21] X86_64, UV: Add Initial UV4 definitions

2016-04-27 Thread Mike Travis
Add preliminary UV4 defines. Signed-off-by: Mike Travis <tra...@sgi.com> Tested-by: John Estabrook <estabr...@sgi.com> Tested-by: Gary Kroening <g...@sgi.com> --- arch/x86/include/asm/uv/uv_hub.h | 64 +++--- arch/x86/include/asm/u

[PATCH 02/21] X86_64, UV: Add UV Architecture Defines

2016-04-27 Thread Mike Travis
Add defines to control which UV architectures are supported, and modify the 'if (is_uvX_*)' functions to return constant 0 for those not supported. This will help optimize code paths when support for specific UV arches is removed. Signed-off-by: Mike Travis <tra...@sgi.com> Reviewed-by: D

[PATCH 07/21] X86_64, UV: Remove Support for UV1

2016-04-27 Thread Mike Travis
The UV1 architecture is no longer being supported in future Linux kernels, so remove UV1 specific code. Signed-off-by: Mike Travis <tra...@sgi.com> Reviewed-by: Dimitri Sivanich <sivan...@sgi.com> Tested-by: John Estabrook <estabr...@sgi.com> Tested-by: Gary Kroening <g...@s

[PATCH 15/21] X86_64, UV: Add UV4 addressing discovery function

2016-04-27 Thread Mike Travis
. Signed-off-by: Mike Travis <tra...@sgi.com> Reviewed-by: Dimitri Sivanich <sivan...@sgi.com> Tested-by: John Estabrook <estabr...@sgi.com> Tested-by: Gary Kroening <g...@sgi.com> --- arch/x86/kernel/apic/x2apic_uv_x.c | 85 +++-- 1 file chang

[PATCH 14/21] X86_64, UV: Fold blade info into per node hub info structs

2016-04-27 Thread Mike Travis
in the comments and whitespace to clean things up a bit. Signed-off-by: Mike Travis <tra...@sgi.com> Tested-by: Dimitri Sivanich <sivan...@sgi.com> Reviewed-by: Andrew Banman <aban...@sgi.com> Tested-by: John Estabrook <estabr...@sgi.com> Tested-by: Gary Kroening <g...@sgi.com>

[PATCH 01/21] X86_64, UV: Add Initial UV4 definitions

2016-04-27 Thread Mike Travis
Add preliminary UV4 defines. Signed-off-by: Mike Travis Tested-by: John Estabrook Tested-by: Gary Kroening --- arch/x86/include/asm/uv/uv_hub.h | 64 +++--- arch/x86/include/asm/uv/uv_mmrs.h | 22 + 2 files changed, 62 insertions(+), 24

[PATCH 02/21] X86_64, UV: Add UV Architecture Defines

2016-04-27 Thread Mike Travis
Add defines to control which UV architectures are supported, and modify the 'if (is_uvX_*)' functions to return constant 0 for those not supported. This will help optimize code paths when support for specific UV arches is removed. Signed-off-by: Mike Travis Reviewed-by: Dimitri Sivanich Tested

[PATCH 07/21] X86_64, UV: Remove Support for UV1

2016-04-27 Thread Mike Travis
The UV1 architecture is no longer being supported in future Linux kernels, so remove UV1 specific code. Signed-off-by: Mike Travis Reviewed-by: Dimitri Sivanich Tested-by: John Estabrook Tested-by: Gary Kroening --- arch/x86/include/asm/uv/uv_hub.h |9 -- arch/x86/kernel/apic

[PATCH 15/21] X86_64, UV: Add UV4 addressing discovery function

2016-04-27 Thread Mike Travis
. Signed-off-by: Mike Travis Reviewed-by: Dimitri Sivanich Tested-by: John Estabrook Tested-by: Gary Kroening --- arch/x86/kernel/apic/x2apic_uv_x.c | 85 +++-- 1 file changed, 72 insertions(+), 13 deletions(-) --- linux.orig/arch/x86/kernel/apic/x2apic_uv_x.c

[PATCH 14/21] X86_64, UV: Fold blade info into per node hub info structs

2016-04-27 Thread Mike Travis
in the comments and whitespace to clean things up a bit. Signed-off-by: Mike Travis Tested-by: Dimitri Sivanich Reviewed-by: Andrew Banman Tested-by: John Estabrook Tested-by: Gary Kroening --- arch/x86/include/asm/uv/uv_hub.h | 71 +--- arch/x86/kernel/apic/x2apic_uv_x.c | 212

[PATCH 04/21] X86_64, UV: Add UV MMR Illegal Access Function

2016-04-27 Thread Mike Travis
the problem. It allows functions common to some UV arches to use common defines that might not be fully defined for all arches, as long as they do not reference them on the unsupported arches. Signed-off-by: Mike Travis <tra...@sgi.com> Reviewed-by: Dimitri Sivanich <sivan...@sgi.co

[PATCH 20/21] X86_64, UV: Remove Obsolete GRU MMR address translation

2016-04-27 Thread Mike Travis
From: Dimitri Sivanich <sivan...@sgi.com> Use no-op messages in place of cross-partition interrupts when nacking a put message in the GRU. This allows us to remove MMR's as a destination from the GRU driver. Signed-off-by: Dimitri Sivanich <sivan...@sgi.com> Signed-off-by: Mike

[PATCH 00/21] X86_64, UV: Update kernel for SGI UV4 support

2016-04-27 Thread Mike Travis
This patch set primarily updates the Linux kernel to support the next generation SGI Ultraviolet system, UV4. This architecture change is a larger incremental change than previous UV updates because of a major change to the addressing scheme. Previous UV architectures used a fixed address width

[PATCH 04/21] X86_64, UV: Add UV MMR Illegal Access Function

2016-04-27 Thread Mike Travis
the problem. It allows functions common to some UV arches to use common defines that might not be fully defined for all arches, as long as they do not reference them on the unsupported arches. Signed-off-by: Mike Travis Reviewed-by: Dimitri Sivanich Tested-by: John Estabrook Tested-by: Gary

[PATCH 20/21] X86_64, UV: Remove Obsolete GRU MMR address translation

2016-04-27 Thread Mike Travis
From: Dimitri Sivanich Use no-op messages in place of cross-partition interrupts when nacking a put message in the GRU. This allows us to remove MMR's as a destination from the GRU driver. Signed-off-by: Dimitri Sivanich Signed-off-by: Mike Travis Tested-by: John Estabrook Tested-by: Gary

[PATCH 00/21] X86_64, UV: Update kernel for SGI UV4 support

2016-04-27 Thread Mike Travis
This patch set primarily updates the Linux kernel to support the next generation SGI Ultraviolet system, UV4. This architecture change is a larger incremental change than previous UV updates because of a major change to the addressing scheme. Previous UV architectures used a fixed address width

[PATCH 10/21] X86_64, UV: Create per cpu info structs to replace per hub info structs

2016-04-27 Thread Mike Travis
The major portion of the hub info is common to all cpus on that hub. This is step one of moving the per cpu hub info to a per node hub info struct. This patch creates the small per cpu info struct that will contain only information specific to each CPU. Signed-off-by: Mike Travis <tra...@sgi.

[PATCH 13/21] X86_64, UV: Allocate common per node hub info structs on local node

2016-04-27 Thread Mike Travis
when a group of cpus on a node are all interrupted for a common task. Signed-off-by: Mike Travis <tra...@sgi.com> Reviewed-by: Dimitri Sivanich <sivan...@sgi.com> Reviewed-by: Andrew Banman <aban...@sgi.com> Tested-by: John Estabrook <estabr...@sgi.com> Tested-by: Gar

[PATCH 10/21] X86_64, UV: Create per cpu info structs to replace per hub info structs

2016-04-27 Thread Mike Travis
The major portion of the hub info is common to all cpus on that hub. This is step one of moving the per cpu hub info to a per node hub info struct. This patch creates the small per cpu info struct that will contain only information specific to each CPU. Signed-off-by: Mike Travis Reviewed

[PATCH 13/21] X86_64, UV: Allocate common per node hub info structs on local node

2016-04-27 Thread Mike Travis
when a group of cpus on a node are all interrupted for a common task. Signed-off-by: Mike Travis Reviewed-by: Dimitri Sivanich Reviewed-by: Andrew Banman Tested-by: John Estabrook Tested-by: Gary Kroening --- arch/x86/include/asm/uv/uv_hub.h | 38 +-- arch/x86/kernel

[PATCH 12/21] X86_64, UV: Move blade local processor ID to the per cpu info struct

2016-04-27 Thread Mike Travis
definition. This allows calling source code to be built for both pre-UV4 kernels as well as post-UV4 kernels. Signed-off-by: Mike Travis <tra...@sgi.com> Reviewed-by: Dimitri Sivanich <sivan...@sgi.com> Tested-by: John Estabrook <estabr...@sgi.com> Tested-by: Gary Kroening <g...@s

<    1   2   3   4   5   6   7   8   9   10   >