x86: remove Summit and Cyclone support
cover x86: remove Summit and Cyclone support
The Cyclone chipset was only used for systems that are now unsupported. These two patches remove this vestigial code.
PATCH 1/2 x86/apic: remove Summit support
IBM's Summit chipset was only used for 32-bit only Intel ccNUMA and IA-64 machines, neither of which are supported by Xen anymore.
--- docs/misc/xen-command-line.markdown | 2 +- xen/arch/x86/genapic/Makefile | 1 - xen/arch/x86/genapic/probe.c | 2 - xen/arch/x86/genapic/summit.c | 27 ------ xen/include/asm-x86/mach-default/mach_mpspec.h | 2 +- xen/include/asm-x86/mach-summit/mach_mpparse.h | 108 ------------------------ 6 files changed, 2 insertions(+), 140 deletions(-) delete mode 100644 xen/arch/x86/genapic/summit.c delete mode 100644 xen/include/asm-x86/mach-summit/mach_mpparse.h
docs/misc/xen-command-line.markdown +1 -1
@@ -132,7 +132,7 @@ domain 0 command line Permit Xen to use superpages when performing memory management. ### apic -> `= summit | bigsmp | default` +> `= bigsmp | default` Override Xen's logic for choosing the APIC driver. By default, if there are more than 8 CPUs, Xen will switch to `bigsmp` over
xen/arch/x86/genapic/Makefile +0 -1
@@ -3,4 +3,3 @@ obj-y += x2apic.o obj-y += default.o obj-y += delivery.o obj-y += probe.o -obj-y += summit.o
xen/arch/x86/genapic/probe.c +0 -2
@@ -16,13 +16,11 @@#include <asm/mach-generic/mach_apic.h> #include <asm/setup.h> -extern const struct genapic apic_summit;extern const struct genapic apic_bigsmp;const struct genapic *__read_mostly genapic;const struct genapic *apic_probe[]__initdata = { - &apic_summit, &apic_bigsmp, &apic_default,/* must be last */ NULL ,
xen/arch/x86/genapic/summit.c deleted +0 -27
@@ -1,27 +0,0 @@ -/* -* APIC driver for the IBM "Summit" chipset. -*/ -#include <xen/config.h> -#include <xen/cpumask.h> -#include <asm/current.h> -#include <asm/mpspec.h> -#include <asm/genapic.h> -#include <asm/fixmap.h> -#include <asm/apicdef.h> -#include <xen/kernel.h> -#include <xen/string.h> -#include <xen/smp.h> -#include <xen/init.h> -#include <asm/mach-summit/mach_mpparse.h> -#include <asm/io_apic.h> - -static __init int probe_summit (void ) -{ -/* probed later in mptable/ACPI hooks */ -return 0 ; -} - -const struct genapic apic_summit = { -APIC_INIT ("summit" , probe_summit), -GENAPIC_PHYS -};
xen/include/asm-x86/mach-default/mach_mpspec.h +1 -1
@@ -3,7 +3,7 @@#define MAX_IRQ_SOURCES 256 -/* Summit or generic (i.e. installer) kernels need lots of bus entries. */ +/* Generic (i.e. installer) kernels need lots of bus entries. */ /* Maximum 256 PCI busses, plus 1 ISA bus in each of 4 cabinets. */ #define MAX_MP_BUSSES 260
xen/include/asm-x86/mach-summit/mach_mpparse.h deleted +0 -108
108 lines not on this page · raw message
Signature · 1 line
1.7.4.5
PATCH 2/2 x86/time: remove Cyclone as a platform timer
In reply to Matt WilsonThe Cyclone time source was part of IBM's Summit chipset, which was only used for 32-bit only ccNUMA and IA-64 machines. Neither of these are supported by Xen anymore.
--- docs/misc/xen-command-line.markdown | 2 +- xen/arch/x86/time.c | 71 +--------------------------------- xen/include/asm-x86/fixmap.h | 1 - 3 files changed, 4 insertions(+), 70 deletions(-)
docs/misc/xen-command-line.markdown +1 -1
@@ -194,7 +194,7 @@ and reboots. If set, override Xen's calculation of the level 2 cache line size. ### clocksource -> `= pit | hpet | cyclone | acpi` +> `= pit | hpet | acpi` If set, override Xen's default choice for the platform timer.
xen/arch/x86/time.c +3 -68
89 lines not on this page · raw message
xen/include/asm-x86/fixmap.h +0 -1
@@ -56,7 +56,6 @@ enum fixed_addresses {FIX_ACPI_BEGIN ,FIX_ACPI_END =FIX_ACPI_BEGIN +FIX_ACPI_PAGES -1 ,FIX_HPET_BASE , -FIX_CYCLONE_TIMER ,FIX_KEXEC_BASE_0 ,FIX_KEXEC_BASE_END =FIX_KEXEC_BASE_0 \ + ((KEXEC_XEN_NO_PAGES >> 1) * KEXEC_IMAGE_NR) - 1,
Signature · 1 line
1.7.4.5
Quoted 3 lines
Quoted 1 line
Quoted 1 line · Matt Wilson
The Cyclone chipset was only used for systems that are now unsupported. These two patches remove this vestigial code.
Is these being used just in 32-bit only systems documented somewhere?
Jan
Quoted 6 lines · Jan Beulich
Quoted 3 lines
Quoted 1 line
Quoted 1 line · Matt Wilson
The Cyclone chipset was only used for systems that are now unsupported. These two patches remove this vestigial code.
Is these being used just in 32-bit only systems documented somewhere?
The Linux kernel config for X86_SUMMIT depends on X86_32_NON_STANDARD, "Support non-standard 32-bit SMP architectures." I worked to enable Summit support in Red Hat's products circa 2001-2002 for IBM's x440.
Regarding support for IA-64, see:
http://www-03.ibm.com/press/us/en/pressrelease/1124.wss
I believe the model for that was x455.
--msw
Quoted 6 lines
Quoted 1 line
Quoted 1 line · Matt Wilson
Quoted 4 lines · Jan Beulich
Quoted 3 lines
Quoted 1 line
Quoted 1 line · Matt Wilson
The Cyclone chipset was only used for systems that are now unsupported. These two patches remove this vestigial code.
Quoted 6 lines
Quoted 2 lines
Is these being used just in 32-bit only systems documented somewhere?
The Linux kernel config for
X86_SUMMITdepends onX86_32_NON_STANDARD, "Support non-standard 32-bit SMP architectures." I worked to enable Summit support in Red Hat's products circa 2001-2002 for IBM's x440.
Ah, yes, okay.
Quoted 4 lines
Regarding support for IA-64, see:
http://www-03.ibm.com/press/us/en/pressrelease/1124.wssI believe the model for that was x455.
IA-64 is of no interest anymore in the context of Xen.
Jan
Quoted 3 lines · Matt Wilson
The Cyclone chipset was only used for systems that are now unsupported. These two patches remove this vestigial code.
It would be more clear if you specifically mentioned that this hardware is from IA-64 servers, which we indeed don't care about any more. (Perhaps tweak the commit messages?)
As for the patches themselves, they do appear to be in order (and thanks for remembering xen-command-line.markdown).
Quoted 8 lines · Andrew Cooper
Quoted 3 lines · Matt Wilson
The Cyclone chipset was only used for systems that are now unsupported. These two patches remove this vestigial code.
It would be more clear if you specifically mentioned that this hardware is from IA-64 servers, which we indeed don't care about any more. (Perhaps tweak the commit messages?)
The individual patch comments say as much. Keep in mind the chipset was used for both 32-bit x86 and IA-64 Intel platforms. Please let me know if there's anything in particular you'd suggest for the patch commit messages.
Quoted 4 lines
As for the patches themselves, they do appear to be in order (and thanks for remembering xen-command-line.markdown).
Reviewed-by Andrew Cooper <[email protected]>
Thanks!
--msw