[PATCH] s390: add kexec support to 2.6.12-rc2-mm3
From: Heiko Carstens
Date: Mon Apr 25 2005 - 06:32:00 EST
Hi Andrew,
the patch below adds kexec support for s390. The patch applies against
2.6.12-rc2-mm3. Could you apply this patch to your -mm tree please?
Thanks,
Heiko
From: Rolf Adelsberger <adelsberger@xxxxxxxxxx>
From: Heiko Carstens <heiko.carstens@xxxxxxxxxx>
Add kexec support for s390 architecture.
Acked-by: Martin Schwidefsky <schwidefsky@xxxxxxxxxx>
Signed-off-by: Heiko Carstens <heiko.carstens@xxxxxxxxxx>
diffstat:
arch/s390/Kconfig | 20 +++++
arch/s390/kernel/Makefile | 10 ++
arch/s390/kernel/compat_wrapper.S | 8 ++
arch/s390/kernel/crash.c | 17 ++++
arch/s390/kernel/machine_kexec.c | 131 +++++++++++++++++++++++++++++++++++
arch/s390/kernel/relocate_kernel.S | 83 ++++++++++++++++++++++
arch/s390/kernel/relocate_kernel64.S | 84 ++++++++++++++++++++++
arch/s390/kernel/syscalls.S | 2
include/asm-s390/kexec.h | 42 +++++++++++
include/asm-s390/unistd.h | 2
include/linux/kexec.h | 1
11 files changed, 398 insertions(+), 2 deletions(-)
diff -urN a/arch/s390/Kconfig b/arch/s390/Kconfig
--- a/arch/s390/Kconfig 2005-04-14 09:50:38.000000000 +0200
+++ b/arch/s390/Kconfig 2005-04-22 13:19:00.000000000 +0200
@@ -455,6 +455,26 @@
The HZ timer is switched off in idle by default. That means the
HZ timer is already disabled at boot time.
+config KEXEC
+ bool "kexec system call (EXPERIMENTAL)"
+ depends on EXPERIMENTAL
+ help
+ kexec is a system call that implements the ability to shutdown your
+ current kernel, and to start another kernel. It is like a reboot
+ but it is indepedent of the system firmware. And like a reboot
+ you can start any kernel with it, not just Linux.
+
+ The name comes from the similiarity to the exec system call.
+
+ It is an ongoing process to be certain the hardware in a machine
+ is properly shutdown, so do not be surprised if this code does not
+ initially work for you. It may help to enable device hotplugging
+ support. As of this writing the exact hardware interface is
+ strongly in flux, so no good recommendation can be made.
+
+ In the GAMECUBE implementation, kexec allows you to load and
+ run DOL files, including kernel and homebrew DOLs.
+
endmenu
config PCMCIA
diff -urN a/arch/s390/kernel/compat_wrapper.S b/arch/s390/kernel/compat_wrapper.S
--- a/arch/s390/kernel/compat_wrapper.S 2005-04-14 09:50:24.000000000 +0200
+++ b/arch/s390/kernel/compat_wrapper.S 2005-04-22 14:13:56.000000000 +0200
@@ -1441,3 +1441,11 @@
lgfr %r5,%r5 # int
llgtr %r6,%r6 # struct rusage_emu31 *
jg compat_sys_waitid
+
+ .globl compat_sys_kexec_load_wrapper
+compat_sys_kexec_load_wrapper:
+ llgfr %r2,%r2 # unsigned long
+ llgfr %r3,%r3 # unsigned long
+ llgtr %r4,%r4 # struct kexec_segment *
+ llgfr %r5,%r5 # unsigned long
+ jg compat_sys_kexec_load
diff -urN a/arch/s390/kernel/crash.c b/arch/s390/kernel/crash.c
--- a/arch/s390/kernel/crash.c 1970-01-01 01:00:00.000000000 +0100
+++ b/arch/s390/kernel/crash.c 2005-04-22 13:45:08.000000000 +0200
@@ -0,0 +1,17 @@
+/*
+ * arch/s390/kernel/crash.c
+ *
+ * (C) Copyright IBM Corp. 2005
+ *
+ * Author(s): Heiko Carstens <heiko.carstens@xxxxxxxxxx>
+ *
+ */
+
+#include <linux/threads.h>
+#include <linux/kexec.h>
+
+note_buf_t crash_notes[NR_CPUS];
+
+void machine_crash_shutdown(void)
+{
+}
diff -urN a/arch/s390/kernel/machine_kexec.c b/arch/s390/kernel/machine_kexec.c
--- a/arch/s390/kernel/machine_kexec.c 1970-01-01 01:00:00.000000000 +0100
+++ b/arch/s390/kernel/machine_kexec.c 2005-04-22 13:44:28.000000000 +0200
@@ -0,0 +1,131 @@
+/*
+ * arch/s390/kernel/machine_kexec.c
+ *
+ * (C) Copyright IBM Corp. 2005
+ *
+ * Author(s): Rolf Adelsberger <adelsberger@xxxxxxxxxx>
+ *
+ */
+
+/*
+ * s390_machine_kexec.c - handle the transition of Linux booting another kernel
+ * on the S390 architecture.
+ */
+
+#include <asm/cio.h>
+#include <asm/setup.h>
+#include <linux/device.h>
+#include <linux/mm.h>
+#include <linux/kexec.h>
+#include <linux/delay.h>
+#include <asm/pgtable.h>
+#include <asm/pgalloc.h>
+#include <asm/system.h>
+
+int machine_kexec_prepare(struct kimage *);
+void machine_kexec_cleanup(struct kimage *);
+void machine_shutdown(void);
+void machine_kexec(struct kimage *);
+static void kexec_halt_all_cpus(void *);
+
+typedef void (*relocate_kernel_t) (unsigned long indirection_page,
+ unsigned long reboot_code_buffer,
+ unsigned long start_address);
+
+const extern unsigned char relocate_kernel[];
+const extern unsigned long long relocate_kernel_len;
+
+/*
+ * Do whatever setup is needed on image and the reboot code buffer to
+ * allow us to aviood allocations later.
+ */
+
+int
+machine_kexec_prepare(struct kimage * image)
+{
+ unsigned long reboot_code_buffer;
+
+ /* Get the destination where the assembler code should be copied to.*/
+ reboot_code_buffer = page_to_pfn(image->control_code_page)<<PAGE_SHIFT;
+
+ /* Then copy it */
+ memcpy((void *) reboot_code_buffer, relocate_kernel,
+ relocate_kernel_len);
+ return 0;
+}
+
+
+/*
+ * We don't have to do anything here, because we didn't have to prepare
+ * (i.e. allocate or the like) anything in 'machine_kexec_prepare'.
+ */
+void
+machine_kexec_cleanup(struct kimage *image)
+{
+}
+
+/*
+ * Just print a message in the kernel log.
+ */
+void
+machine_shutdown(void)
+{
+ printk(KERN_INFO "kexec: machine_shutdown called\n");
+}
+
+/*
+ * The addresses for the reboot_code_buffer, the indirection page and the
+ * data_mover can already be assigned here.
+ * Also, we can here clear all subchannels.
+ * In any case we have to disable the lowcore protection, because we will
+ * write a new PSW to address 0.
+ * Then we call the function 'kexec_halt_all_cpus'.
+ */
+
+static unsigned long indirection_page;
+static unsigned long reboot_code_buffer;
+static relocate_kernel_t data_mover;
+
+NORET_TYPE void
+machine_kexec(struct kimage *image)
+{
+ reboot_code_buffer = page_to_pfn(image->control_code_page)<<PAGE_SHIFT;
+ indirection_page = image->head & PAGE_MASK;
+ data_mover = (relocate_kernel_t) reboot_code_buffer;
+
+ clear_all_subchannels();
+
+ /* Disable lowcore protection */
+ ctl_clear_bit(0,28);
+
+ on_each_cpu(kexec_halt_all_cpus, image, 0, 0);
+ for(;;);
+}
+
+/*
+ * Make sure that all except one CPU are stopped. The first CPU
+ * that enters this function is going to halt all the others and then, if it is
+ * the only one left, the 'data_mover' function will be called.
+ */
+static void
+kexec_halt_all_cpus(void *kernel_image)
+{
+ static atomic_t cpuid = ATOMIC_INIT(-1);
+ int cpu;
+
+ struct kimage *image = (struct kimage *) kernel_image;
+
+ if(atomic_compare_and_swap(-1, smp_processor_id(), &cpuid))
+ signal_processor(smp_processor_id(), sigp_stop);
+
+ /* Wait for all other cpus to enter stopped state */
+ for_each_online_cpu(cpu) {
+ if (cpu == smp_processor_id())
+ continue;
+ while(!smp_cpu_not_running(cpu))
+ cpu_relax();
+ }
+
+ /* Call the moving routine */
+ (*data_mover) (indirection_page, reboot_code_buffer, image->start);
+}
diff -urN a/arch/s390/kernel/Makefile b/arch/s390/kernel/Makefile
--- a/arch/s390/kernel/Makefile 2005-03-02 08:38:32.000000000 +0100
+++ b/arch/s390/kernel/Makefile 2005-04-22 13:19:00.000000000 +0200
@@ -25,6 +25,16 @@
obj-$(CONFIG_VIRT_TIMER) += vtime.o
+# Kexec part
+S390_KEXEC_OBJS := machine_kexec.o crash.o
+ifeq ($(CONFIG_ARCH_S390X),y)
+S390_KEXEC_OBJS += relocate_kernel64.o
+else
+S390_KEXEC_OBJS += relocate_kernel.o
+endif
+obj-$(CONFIG_KEXEC) += $(S390_KEXEC_OBJS)
+
+
#
# This is just to get the dependencies...
#
diff -urN a/arch/s390/kernel/relocate_kernel64.S b/arch/s390/kernel/relocate_kernel64.S
--- a/arch/s390/kernel/relocate_kernel64.S 1970-01-01 01:00:00.000000000 +0100
+++ b/arch/s390/kernel/relocate_kernel64.S 2005-04-22 13:43:07.000000000 +0200
@@ -0,0 +1,84 @@
+/*
+ * arch/s390/kernel/relocate_kernel64.S
+ *
+ * (C) Copyright IBM Corp. 2005
+ *
+ * Author(s): Rolf Adelsberger <adelsberger@xxxxxxxxxx>
+ *
+ */
+
+/*
+ * moves the new kernel to its destination...
+ * %r2 = indirection page
+ * %r3 = reboot_code_buffer
+ * %r4 = start address - where to jump to after the job is done...
+ *
+ * %r5 will be used as temp. storage
+ * %r6 holds the destination address
+ * %r7 = PAGE_SIZE
+ * %r8 holds the source address
+ * %r9 = PAGE_SIZE
+ *
+ * 0xf000 is a page_mask
+ */
+
+ .text
+ .globl relocate_kernel
+ relocate_kernel:
+ basr %r13,0 #base address
+ .base:
+ spx zero64-.base(%r13) #absolute addressing mode
+ stnsm sys_msk-.base(%r13),0xf8 #disable DAT and IRQ (external)
+ .top:
+ lghi %r7,4096 #load PAGE_SIZE in r7
+ lghi %r9,4096 #load PAGE_SIZE in r9
+ lg %r5,0(%r2) #read another word for indirection page
+ aghi %r2,8 #increment pointer
+ tml %r5,0x1 #is it a destination page?
+ je .indir_check #NO, goto "indir_check"
+ lgr %r6,%r5 #r6 = r5
+ nill %r6,0xf000 #mask it out and...
+ j .top #...next iteration
+ .indir_check:
+ tml %r5,0x2 #is it a indirection page?
+ je .done_test #NO, goto "done_test"
+ nill %r5,0xf000 #YES, mask out,
+ lgr %r2,%r5 #move it into the right register,
+ j .top #and read next...
+ .done_test:
+ tml %r5,0x4 #is it the done indicator?
+ je .source_test #NO! Well, then it should be the source indicator...
+ j .done #ok, lets finish it here...
+ .source_test:
+ tml %r5,0x8 #it should be a source indicator...
+ je .top #NO, ignore it...
+ lgr %r8,%r5 #r8 = r5
+ nill %r8,0xf000 #masking
+ 0: mvcle %r6,%r8,0x0 #copy PAGE_SIZE bytes from r8 to r6 - pad with 0
+ jo 0b
+ j .top
+ .done:
+ sgr %r0,%r0 #clear register r0
+ la %r3,load_psw-.base(%r13) #load psw-address into the register
+ mvc 0(4,%r0),0(%r3) #copy the first part of it
+ l %r1,4(0,%r3) #copy the second part of load_psw in %r1
+ or %r1,%r4 #OR it with the given psw --> %r1 = (%r1 | %r4)
+ st %r1,4(0,%r0) #store it at addr 0+4B, i.e. second part of psw
+ sam31 #31 bit mode
+ sr %r1,%r1 #erase register r1
+ sr %r2,%r2 #erase register r2
+ sigp %r1,%r2,0x12 #set cpuid to zero
+ lpsw 0 #hopefully start new kernel...
+
+ .align 8
+ zero64:
+ .quad 0
+ load_psw:
+ .long 0x00080000,0x80000000
+ sys_msk:
+ .quad 0
+ relocate_kernel_end:
+ .globl relocate_kernel_len
+ relocate_kernel_len:
+ .quad relocate_kernel_end - relocate_kernel
+
diff -urN a/arch/s390/kernel/relocate_kernel.S b/arch/s390/kernel/relocate_kernel.S
--- a/arch/s390/kernel/relocate_kernel.S 1970-01-01 01:00:00.000000000 +0100
+++ b/arch/s390/kernel/relocate_kernel.S 2005-04-22 13:40:52.000000000 +0200
@@ -0,0 +1,83 @@
+/*
+ * arch/s390/kernel/relocate_kernel.S
+ *
+ * (C) Copyright IBM Corp. 2005
+ *
+ * Author(s): Rolf Adelsberger <adelsberger@xxxxxxxxxx>
+ *
+ */
+
+/*
+ * moves the new kernel to its destination...
+ * %r2 = indirection page
+ * %r3 = reboot_code_buffer
+ * %r4 = start address - where to jump to after the job is done...
+ *
+ * %r5 will be used as temp. storage
+ * %r6 holds the destination address
+ * %r7 = PAGE_SIZE
+ * %r8 holds the source address
+ * %r9 = PAGE_SIZE
+ * %r10 is a page mask
+ */
+
+ .text
+ .globl relocate_kernel
+ relocate_kernel:
+ basr %r13,0 #base address
+ .base:
+ spx zero64-.base(%r13) #absolute addressing mode
+ stnsm sys_msk-.base(%r13),0xf8 #disable DAT and IRQ (external)
+ lhi %r10,-1 #preparing the mask
+ sll %r10,12 #shift it such that it becomes 0xf000
+ .top:
+ lhi %r7,4096 #load PAGE_SIZE in r7
+ lhi %r9,4096 #load PAGE_SIZE in r9
+ l %r5,0(%r2) #read another word for indirection page
+ ahi %r2,4 #increment pointer
+ tml %r5,0x1 #is it a destination page?
+ je .indir_check #NO, goto "indir_check"
+ lr %r6,%r5 #r6 = r5
+ nr %r6,%r10 #mask it out and...
+ j .top #...next iteration
+ .indir_check:
+ tml %r5,0x2 #is it a indirection page?
+ je .done_test #NO, goto "done_test"
+ nr %r5,%r10 #YES, mask out,
+ lr %r2,%r5 #move it into the right register,
+ j .top #and read next...
+ .done_test:
+ tml %r5,0x4 #is it the done indicator?
+ je .source_test #NO! Well, then it should be the source indicator...
+ j .done #ok, lets finish it here...
+ .source_test:
+ tml %r5,0x8 #it should be a source indicator...
+ je .top #NO, ignore it...
+ lr %r8,%r5 #r8 = r5
+ nr %r8,%r10 #masking
+ 0: mvcle %r6,%r8,0x0 #copy PAGE_SIZE bytes from r8 to r6 - pad with 0
+ jo 0b
+ j .top
+ .done:
+ sr %r0,%r0 #clear register r0
+ la %r3,load_psw-.base(%r13) #load psw-address into the register
+ mvc 0(4,%r0),0(%r3) #copy the first part of psw
+ l %r1,4(0,%r3) #get the second part of psw in %r1
+ or %r1,%r4 #OR the load address with the 2nd part of psw
+ st %r1,4(0,%r0) #store it at add 0+4B
+ sr %r1,%r1 #clear %r1
+ sr %r2,%r2 #clear %r2
+ sigp %r1,%r2,0x12 #set cpuid to zero
+ lpsw 0 #hopefully start new kernel...
+
+ .align 8
+ zero64:
+ .quad 0
+ load_psw:
+ .long 0x00080000,0x80000000
+ sys_msk:
+ .quad 0
+ relocate_kernel_end:
+ .globl relocate_kernel_len
+ relocate_kernel_len:
+ .quad relocate_kernel_end - relocate_kernel
diff -urN a/arch/s390/kernel/syscalls.S b/arch/s390/kernel/syscalls.S
--- a/arch/s390/kernel/syscalls.S 2005-04-14 09:50:24.000000000 +0200
+++ b/arch/s390/kernel/syscalls.S 2005-04-22 14:12:04.000000000 +0200
@@ -285,7 +285,7 @@
SYSCALL(sys_mq_timedreceive,sys_mq_timedreceive,compat_sys_mq_timedreceive_wrapper)
SYSCALL(sys_mq_notify,sys_mq_notify,compat_sys_mq_notify_wrapper) /* 275 */
SYSCALL(sys_mq_getsetattr,sys_mq_getsetattr,compat_sys_mq_getsetattr_wrapper)
-NI_SYSCALL /* reserved for kexec */
+SYSCALL(sys_kexec_load,sys_kexec_load,compat_sys_kexec_load_wrapper)
SYSCALL(sys_add_key,sys_add_key,compat_sys_add_key_wrapper)
SYSCALL(sys_request_key,sys_request_key,compat_sys_request_key_wrapper)
SYSCALL(sys_keyctl,sys_keyctl,compat_sys_keyctl) /* 280 */
diff -urN a/include/asm-s390/kexec.h b/include/asm-s390/kexec.h
--- a/include/asm-s390/kexec.h 1970-01-01 01:00:00.000000000 +0100
+++ b/include/asm-s390/kexec.h 2005-04-22 13:47:29.000000000 +0200
@@ -0,0 +1,42 @@
+/*
+ * include/asm-s390/kexec.h
+ *
+ * (C) Copyright IBM Corp. 2005
+ *
+ * Author(s): Rolf Adelsberger <adelsberger@xxxxxxxxxx>
+ *
+ */
+
+#ifndef _S390_KEXEC_H
+#define _S390_KEXEC_H
+
+#include <asm/page.h>
+#include <asm/processor.h>
+/*
+ * KEXEC_SOURCE_MEMORY_LIMIT maximum page get_free_page can return.
+ * I.e. Maximum page that is mapped directly into kernel memory,
+ * and kmap is not required.
+ */
+
+/* Maximum physical address we can use pages from */
+#define KEXEC_SOURCE_MEMORY_LIMIT (-1UL)
+
+/* Maximum address we can reach in physical address mode */
+#define KEXEC_DESTINATION_MEMORY_LIMIT (-1UL)
+
+/* Maximum address we can use for the control pages */
+/* Not more than 2GB */
+#define KEXEC_CONTROL_MEMORY_LIMIT (1<<31)
+
+/* Allocate one page for the pdp and the second for the code */
+#define KEXEC_CONTROL_CODE_SIZE 4096
+
+/* The native architecture */
+#define KEXEC_ARCH KEXEC_ARCH_S390
+
+#define MAX_NOTE_BYTES 1024
+typedef u32 note_buf_t[MAX_NOTE_BYTES/4];
+
+extern note_buf_t crash_notes[];
+
+#endif /*_S390_KEXEC_H */
diff -urN a/include/asm-s390/unistd.h b/include/asm-s390/unistd.h
--- a/include/asm-s390/unistd.h 2005-04-14 09:50:32.000000000 +0200
+++ b/include/asm-s390/unistd.h 2005-04-22 13:19:00.000000000 +0200
@@ -269,7 +269,7 @@
#define __NR_mq_timedreceive 274
#define __NR_mq_notify 275
#define __NR_mq_getsetattr 276
-/* Number 277 is reserved for new sys_kexec_load */
+#define __NR_kexec_load 277
#define __NR_add_key 278
#define __NR_request_key 279
#define __NR_keyctl 280
diff -urN a/include/linux/kexec.h b/include/linux/kexec.h
--- a/include/linux/kexec.h 2005-04-14 09:50:42.000000000 +0200
+++ b/include/linux/kexec.h 2005-04-22 13:19:00.000000000 +0200
@@ -115,6 +115,7 @@
#define KEXEC_ARCH_PPC (20 << 16)
#define KEXEC_ARCH_PPC64 (21 << 16)
#define KEXEC_ARCH_IA_64 (50 << 16)
+#define KEXEC_ARCH_S390 (22 << 16)
#define KEXEC_FLAGS (KEXEC_ON_CRASH) /* List of defined/legal kexec flags */
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/