* [PATCH v5 33/36] Hexagon: Add configuration and makefiles for the Hexagon architecture.
From: Richard Kuo @ 2011-10-31 23:55 UTC (permalink / raw)
To: torvalds, linux-kernel
Signed-off-by: Linas Vepstas <linas@codeaurora.org>
Signed-off-by: Richard Kuo <rkuo@codeaurora.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
---
arch/hexagon/Kconfig | 220 +++++++++++++++++++++++++++++++++++++
arch/hexagon/Makefile | 59 ++++++++++
arch/hexagon/kernel/Makefile | 19 +++
arch/hexagon/kernel/asm-offsets.c | 104 +++++++++++++++++
arch/hexagon/kernel/vmlinux.lds.S | 93 ++++++++++++++++
arch/hexagon/lib/Makefile | 4 +
arch/hexagon/mm/Makefile | 7 +
7 files changed, 506 insertions(+), 0 deletions(-)
create mode 100644 arch/hexagon/Kconfig
create mode 100644 arch/hexagon/Makefile
create mode 100644 arch/hexagon/kernel/Makefile
create mode 100644 arch/hexagon/kernel/asm-offsets.c
create mode 100644 arch/hexagon/kernel/vmlinux.lds.S
create mode 100644 arch/hexagon/lib/Makefile
create mode 100644 arch/hexagon/mm/Makefile
diff --git a/arch/hexagon/Kconfig b/arch/hexagon/Kconfig
new file mode 100644
index 0000000..02513c2
--- /dev/null
+++ b/arch/hexagon/Kconfig
@@ -0,0 +1,220 @@
+# Hexagon configuration
+comment "Linux Kernel Configuration for Hexagon"
+
+config HEXAGON
+ def_bool y
+ select HAVE_OPROFILE
+ select USE_GENERIC_SMP_HELPERS if SMP
+ # Other pending projects/to-do items.
+ # select HAVE_REGS_AND_STACK_ACCESS_API
+ # select HAVE_HW_BREAKPOINT if PERF_EVENTS
+ # select ARCH_HAS_CPU_IDLE_WAIT
+ # select ARCH_WANT_OPTIONAL_GPIOLIB
+ # select ARCH_REQUIRE_GPIOLIB
+ # select HAVE_CLK
+ # select IRQ_PER_CPU
+ select HAVE_IRQ_WORK
+ # select GENERIC_PENDING_IRQ if SMP
+ select GENERIC_ATOMIC64
+ select HAVE_PERF_EVENTS
+ select HAVE_GENERIC_HARDIRQS
+ select GENERIC_HARDIRQS_NO__DO_IRQ
+ select GENERIC_HARDIRQS_NO_DEPRECATED
+ # GENERIC_ALLOCATOR is used by dma_alloc_coherent()
+ select GENERIC_ALLOCATOR
+ select GENERIC_IRQ_SHOW
+ select HAVE_ARCH_KGDB
+ select HAVE_ARCH_TRACEHOOK
+ select NO_IOPORT
+ # mostly generic routines, with some accelerated ones
+ ---help---
+ Qualcomm Hexagon is a processor architecture designed for high
+ performance and low power across a wide variety of applications.
+
+config HEXAGON_ARCH_V1
+ bool
+
+config HEXAGON_ARCH_V2
+ bool
+
+config HEXAGON_ARCH_V3
+ bool
+
+config HEXAGON_ARCH_V4
+ bool
+
+config FRAME_POINTER
+ def_bool y
+
+config LOCKDEP_SUPPORT
+ def_bool y
+
+config PCI
+ def_bool n
+
+config EARLY_PRINTK
+ def_bool y
+
+config KTIME_SCALAR
+ def_bool y
+
+config MMU
+ def_bool y
+
+config TRACE_IRQFLAGS_SUPPORT
+ def_bool y
+
+config GENERIC_CSUM
+ def_bool y
+
+#
+# Use the generic interrupt handling code in kernel/irq/:
+#
+config GENERIC_IRQ_PROBE
+ def_bool y
+
+config GENERIC_IOMAP
+ def_bool y
+
+#config ZONE_DMA
+# bool
+# default y
+
+config HAS_DMA
+ bool
+ select HAVE_DMA_ATTRS
+ default y
+
+config NEED_SG_DMA_LENGTH
+ def_bool y
+
+config RWSEM_GENERIC_SPINLOCK
+ def_bool n
+
+config RWSEM_XCHGADD_ALGORITHM
+ def_bool y
+
+config GENERIC_FIND_NEXT_BIT
+ def_bool y
+
+config GENERIC_HWEIGHT
+ def_bool y
+
+config GENERIC_TIME
+ def_bool y
+
+config GENERIC_CLOCKEVENTS
+ def_bool y
+
+config GENERIC_CLOCKEVENTS_BROADCAST
+ def_bool y
+
+config STACKTRACE_SUPPORT
+ def_bool y
+ select STACKTRACE
+
+config GENERIC_BUG
+ def_bool y
+ depends on BUG
+
+config BUG
+ def_bool y
+
+menu "Machine selection"
+
+choice
+ prompt "System type"
+ default HEXAGON_ARCH_V2
+
+config HEXAGON_COMET
+ bool "Comet Board"
+ select HEXAGON_ARCH_V2
+ ---help---
+ Support for the Comet platform.
+
+endchoice
+
+config HEXAGON_VM
+ def_bool y
+
+config CMDLINE
+ string "Default kernel command string"
+ default ""
+ help
+ On some platforms, there is currently no way for the boot loader
+ to pass arguments to the kernel. For these, you should supply some
+ command-line options at build time by entering them here. At a
+ minimum, you should specify the memory size and the root device
+ (e.g., mem=64M root=/dev/nfs).
+
+config HEXAGON_ANGEL_TRAPS
+ bool "Use Angel Traps"
+ default n
+ ---help---
+ Enable angel debug traps (for printk's).
+
+config SMP
+ bool "Multi-Processing support"
+ ---help---
+ Enables SMP support in the kernel. If unsure, say "Y"
+
+config NR_CPUS
+ int "Maximum number of CPUs" if SMP
+ range 2 6 if SMP
+ default "1" if !SMP
+ default "6" if SMP
+ ---help---
+ This allows you to specify the maximum number of CPUs which this
+ kernel will support. The maximum supported value is 6 and the
+ minimum value which makes sense is 2.
+
+ This is purely to save memory - each supported CPU adds
+ approximately eight kilobytes to the kernel image.
+
+choice
+ prompt "Kernel page size"
+ default PAGE_SIZE_4KB
+ ---help---
+ Changes the default page size; use with caution.
+
+config PAGE_SIZE_4KB
+ bool "4KB"
+
+config PAGE_SIZE_16KB
+ bool "16KB"
+
+config PAGE_SIZE_64KB
+ bool "64KB"
+
+config PAGE_SIZE_256KB
+ bool "256KB"
+
+endchoice
+
+source "mm/Kconfig"
+
+source "kernel/Kconfig.hz"
+source "kernel/time/Kconfig"
+
+config GENERIC_GPIO
+ bool "Generic GPIO support"
+ default n
+
+endmenu
+
+source "init/Kconfig"
+source "drivers/Kconfig"
+source "fs/Kconfig"
+
+menu "Executable File Formats"
+source "fs/Kconfig.binfmt"
+endmenu
+
+source "net/Kconfig"
+source "security/Kconfig"
+source "crypto/Kconfig"
+source "lib/Kconfig"
+
+menu "Kernel hacking"
+source "lib/Kconfig.debug"
+endmenu
diff --git a/arch/hexagon/Makefile b/arch/hexagon/Makefile
new file mode 100644
index 0000000..7ce9563
--- /dev/null
+++ b/arch/hexagon/Makefile
@@ -0,0 +1,59 @@
+# Makefile for the Hexagon arch
+
+KBUILD_DEFCONFIG = comet_defconfig
+
+# Do not use GP-relative jumps
+KBUILD_CFLAGS += -G0
+LDFLAGS_vmlinux += -G0
+
+# Do not use single-byte enums; these will overflow.
+KBUILD_CFLAGS += -fno-short-enums
+
+# Modules must use either long-calls, or use pic/plt.
+# Use long-calls for now, it's easier. And faster.
+# CFLAGS_MODULE += -fPIC
+# LDFLAGS_MODULE += -shared
+CFLAGS_MODULE += -mlong-calls
+
+cflags-$(CONFIG_HEXAGON_ARCH_V1) += $(call cc-option,-mv1)
+cflags-$(CONFIG_HEXAGON_ARCH_V2) += $(call cc-option,-mv2)
+cflags-$(CONFIG_HEXAGON_ARCH_V3) += $(call cc-option,-mv3)
+cflags-$(CONFIG_HEXAGON_ARCH_V4) += $(call cc-option,-mv4)
+
+aflags-$(CONFIG_HEXAGON_ARCH_V1) += $(call cc-option,-mv1)
+aflags-$(CONFIG_HEXAGON_ARCH_V2) += $(call cc-option,-mv2)
+aflags-$(CONFIG_HEXAGON_ARCH_V3) += $(call cc-option,-mv3)
+aflags-$(CONFIG_HEXAGON_ARCH_V4) += $(call cc-option,-mv4)
+
+ldflags-$(CONFIG_HEXAGON_ARCH_V1) += $(call cc-option,-mv1)
+ldflags-$(CONFIG_HEXAGON_ARCH_V2) += $(call cc-option,-mv2)
+ldflags-$(CONFIG_HEXAGON_ARCH_V3) += $(call cc-option,-mv3)
+ldflags-$(CONFIG_HEXAGON_ARCH_V4) += $(call cc-option,-mv4)
+
+KBUILD_CFLAGS += $(cflags-y)
+KBUILD_AFLAGS += $(aflags-y)
+
+# no KBUILD_LDFLAGS?
+LDFLAGS += $(ldflags-y)
+
+# Thread-info register will be r19. This value is not configureable;
+# it is hard-coded in several files.
+TIR_NAME := r19
+KBUILD_CFLAGS += -ffixed-$(TIR_NAME) -DTHREADINFO_REG=$(TIR_NAME) -D__linux__
+KBUILD_AFLAGS += -DTHREADINFO_REG=$(TIR_NAME)
+
+LIBGCC := $(shell $(CC) $(KBUILD_CFLAGS) -print-libgcc-file-name)
+libs-y += $(LIBGCC)
+
+head-y := arch/hexagon/kernel/head.o \
+ arch/hexagon/kernel/init_task.o
+
+core-y += arch/hexagon/kernel/ \
+ arch/hexagon/mm/ \
+ arch/hexagon/lib/
+
+# arch/hexagon/platform/common/
+#
+#core-$(CONFIG_HEXAGON_COMET) += arch/hexagon/platform/comet/
+#machine-$(CONFIG_HEXAGON_COMET) := comet
+
diff --git a/arch/hexagon/kernel/Makefile b/arch/hexagon/kernel/Makefile
new file mode 100644
index 0000000..27be2f6
--- /dev/null
+++ b/arch/hexagon/kernel/Makefile
@@ -0,0 +1,19 @@
+extra-y := head.o vmlinux.lds init_task.o
+
+obj-$(CONFIG_SMP) += smp.o topology.o
+
+obj-y += setup.o irq_cpu.o traps.o syscalltab.o signal.o time.o
+obj-y += process.o syscall.o trampoline.o reset.o ptrace.o
+obj-y += vdso.o
+
+obj-$(CONFIG_KGDB) += kgdb.o
+obj-$(CONFIG_MODULES) += module.o hexagon_ksyms.o
+
+# Modules required to work with the Hexagon Virtual Machine
+obj-y += vm_entry.o vm_events.o vm_switch.o vm_ops.o vm_init_segtable.o
+obj-y += vm_vectors.o
+
+obj-$(CONFIG_HAS_DMA) += dma.o
+
+obj-$(CONFIG_STACKTRACE) += stacktrace.o
+
diff --git a/arch/hexagon/kernel/asm-offsets.c b/arch/hexagon/kernel/asm-offsets.c
new file mode 100644
index 0000000..89ffa51
--- /dev/null
+++ b/arch/hexagon/kernel/asm-offsets.c
@@ -0,0 +1,104 @@
+/*
+ * Copyright (C) 1996 David S. Miller
+ * Copyright (C) 1997, 1998, 1999, 2000, 2001, 2002, 2003 Ralf Baechle
+ * Copyright (C) 1999, 2000 Silicon Graphics, Inc.
+ * Kevin Kissell, kevink@mips.com and Carsten Langgaard, carstenl@mips.com
+ * Copyright (C) 2000 MIPS Technologies, Inc.
+ *
+ * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+#include <linux/compat.h>
+#include <linux/types.h>
+#include <linux/sched.h>
+#include <linux/mm.h>
+#include <linux/interrupt.h>
+#include <linux/kbuild.h>
+#include <asm/ptrace.h>
+#include <asm/processor.h>
+
+/* This file is used to produce asm/linkerscript constants from header
+ files typically used in c. Specifically, it generates asm-offsets.h */
+
+int main(void)
+{
+ COMMENT("This is a comment.");
+ /* might get these from somewhere else. */
+ DEFINE(_PAGE_SIZE, PAGE_SIZE);
+ DEFINE(_PAGE_SHIFT, PAGE_SHIFT);
+ BLANK();
+
+ COMMENT("Hexagon pt_regs definitions");
+ OFFSET(_PT_SYSCALL_NR, pt_regs, syscall_nr);
+ OFFSET(_PT_UGPGP, pt_regs, ugpgp);
+ OFFSET(_PT_R3130, pt_regs, r3130);
+ OFFSET(_PT_R2928, pt_regs, r2928);
+ OFFSET(_PT_R2726, pt_regs, r2726);
+ OFFSET(_PT_R2524, pt_regs, r2524);
+ OFFSET(_PT_R2322, pt_regs, r2322);
+ OFFSET(_PT_R2120, pt_regs, r2120);
+ OFFSET(_PT_R1918, pt_regs, r1918);
+ OFFSET(_PT_R1716, pt_regs, r1716);
+ OFFSET(_PT_R1514, pt_regs, r1514);
+ OFFSET(_PT_R1312, pt_regs, r1312);
+ OFFSET(_PT_R1110, pt_regs, r1110);
+ OFFSET(_PT_R0908, pt_regs, r0908);
+ OFFSET(_PT_R0706, pt_regs, r0706);
+ OFFSET(_PT_R0504, pt_regs, r0504);
+ OFFSET(_PT_R0302, pt_regs, r0302);
+ OFFSET(_PT_R0100, pt_regs, r0100);
+ OFFSET(_PT_LC0SA0, pt_regs, lc0sa0);
+ OFFSET(_PT_LC1SA1, pt_regs, lc1sa1);
+ OFFSET(_PT_M1M0, pt_regs, m1m0);
+ OFFSET(_PT_PREDSUSR, pt_regs, predsusr);
+ OFFSET(_PT_EVREC, pt_regs, hvmer);
+ OFFSET(_PT_ER_VMEL, pt_regs, hvmer.vmel);
+ OFFSET(_PT_ER_VMEST, pt_regs, hvmer.vmest);
+ OFFSET(_PT_ER_VMPSP, pt_regs, hvmer.vmpsp);
+ OFFSET(_PT_ER_VMBADVA, pt_regs, hvmer.vmbadva);
+ DEFINE(_PT_REGS_SIZE, sizeof(struct pt_regs));
+ BLANK();
+
+ COMMENT("Hexagon thread_info definitions");
+ OFFSET(_THREAD_INFO_FLAGS, thread_info, flags);
+ OFFSET(_THREAD_INFO_PT_REGS, thread_info, regs);
+ OFFSET(_THREAD_INFO_SP, thread_info, sp);
+ DEFINE(_THREAD_SIZE, THREAD_SIZE);
+ BLANK();
+
+ COMMENT("Hexagon hexagon_switch_stack definitions");
+ OFFSET(_SWITCH_R1716, hexagon_switch_stack, r1716);
+ OFFSET(_SWITCH_R1918, hexagon_switch_stack, r1918);
+ OFFSET(_SWITCH_R2120, hexagon_switch_stack, r2120);
+ OFFSET(_SWITCH_R2322, hexagon_switch_stack, r2322);
+
+ OFFSET(_SWITCH_R2524, hexagon_switch_stack, r2524);
+ OFFSET(_SWITCH_R2726, hexagon_switch_stack, r2726);
+ OFFSET(_SWITCH_FP, hexagon_switch_stack, fp);
+ OFFSET(_SWITCH_LR, hexagon_switch_stack, lr);
+ DEFINE(_SWITCH_STACK_SIZE, sizeof(struct hexagon_switch_stack));
+ BLANK();
+
+ COMMENT("Hexagon task_struct definitions");
+ OFFSET(_TASK_THREAD_INFO, task_struct, stack);
+ OFFSET(_TASK_STRUCT_THREAD, task_struct, thread);
+
+ COMMENT("Hexagon thread_struct definitions");
+ OFFSET(_THREAD_STRUCT_SWITCH_SP, thread_struct, switch_sp);
+
+ return 0;
+}
diff --git a/arch/hexagon/kernel/vmlinux.lds.S b/arch/hexagon/kernel/vmlinux.lds.S
new file mode 100644
index 0000000..362742f
--- /dev/null
+++ b/arch/hexagon/kernel/vmlinux.lds.S
@@ -0,0 +1,93 @@
+/*
+ * Linker script for Hexagon kernel
+ *
+ * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+#define LOAD_OFFSET PAGE_OFFSET
+
+#include <asm-generic/vmlinux.lds.h>
+#include <asm/asm-offsets.h> /* Most of the kernel defines are here */
+#include <asm/mem-layout.h> /* except for page_offset */
+#include <asm/cache.h> /* and now we're pulling cache line size */
+OUTPUT_ARCH(hexagon)
+ENTRY(stext)
+
+jiffies = jiffies_64;
+
+/*
+See asm-generic/vmlinux.lds.h for expansion of some of these macros.
+See asm-generic/sections.h for seemingly required labels.
+*/
+
+#define PAGE_SIZE _PAGE_SIZE
+
+/* This LOAD_OFFSET is temporary for debugging on the simulator; it may change
+ for hypervisor pseudo-physical memory. */
+
+
+SECTIONS
+{
+ . = PAGE_OFFSET + LOAD_ADDRESS;
+
+ __init_begin = .;
+ HEAD_TEXT_SECTION
+ INIT_TEXT_SECTION(PAGE_SIZE)
+ PERCPU_SECTION(L1_CACHE_BYTES)
+ __init_end = .;
+
+ . = ALIGN(_PAGE_SIZE);
+ _stext = .;
+ .text : AT(ADDR(.text) - LOAD_OFFSET) {
+ _text = .;
+ TEXT_TEXT
+ SCHED_TEXT
+ LOCK_TEXT
+ KPROBES_TEXT
+ *(.fixup)
+ }
+ _etext = .;
+
+ INIT_DATA_SECTION(PAGE_SIZE)
+
+ _sdata = .;
+ RW_DATA_SECTION(32,PAGE_SIZE,PAGE_SIZE)
+ RO_DATA_SECTION(PAGE_SIZE)
+ _edata = .;
+
+ EXCEPTION_TABLE(16)
+ NOTES
+
+ BSS_SECTION(_PAGE_SIZE, _PAGE_SIZE, _PAGE_SIZE)
+
+ _end = .;
+
+ /DISCARD/ : {
+ EXIT_TEXT
+ EXIT_DATA
+ EXIT_CALL
+ }
+
+ STABS_DEBUG
+ DWARF_DEBUG
+
+}
+
+
+
+
+
diff --git a/arch/hexagon/lib/Makefile b/arch/hexagon/lib/Makefile
new file mode 100644
index 0000000..874655e
--- /dev/null
+++ b/arch/hexagon/lib/Makefile
@@ -0,0 +1,4 @@
+#
+# Makefile for hexagon-specific library files.
+#
+obj-y = checksum.o io.o memcpy.o memset.o
diff --git a/arch/hexagon/mm/Makefile b/arch/hexagon/mm/Makefile
new file mode 100644
index 0000000..018be76
--- /dev/null
+++ b/arch/hexagon/mm/Makefile
@@ -0,0 +1,7 @@
+#
+# Makefile for Hexagon memory management subsystem
+#
+
+obj-y := init.o pgalloc.o ioremap.o uaccess.o vm_fault.o cache.o
+obj-y += copy_to_user.o copy_from_user.o strnlen_user.o vm_tlb.o
+
--
1.7.1
--
Sent by an employee of the Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum.
^ permalink raw reply related
* [PATCH v5 34/36] Hexagon: Add basic stacktrace functionality for Hexagon architecture.
From: Richard Kuo @ 2011-10-31 23:56 UTC (permalink / raw)
To: torvalds, linux-kernel
Signed-off-by: Richard Kuo <rkuo@codeaurora.org>
Signed-off-by: Linas Vepstas <linas@codeaurora.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
---
arch/hexagon/kernel/stacktrace.c | 66 ++++++++++++++++++++++++++++++++++++++
1 files changed, 66 insertions(+), 0 deletions(-)
create mode 100644 arch/hexagon/kernel/stacktrace.c
diff --git a/arch/hexagon/kernel/stacktrace.c b/arch/hexagon/kernel/stacktrace.c
new file mode 100644
index 0000000..11c597b
--- /dev/null
+++ b/arch/hexagon/kernel/stacktrace.c
@@ -0,0 +1,66 @@
+/*
+ * Stacktrace support for Hexagon
+ *
+ * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+#include <linux/sched.h>
+#include <linux/stacktrace.h>
+#include <linux/thread_info.h>
+#include <linux/module.h>
+
+register unsigned long current_frame_pointer asm("r30");
+
+struct stackframe {
+ unsigned long fp;
+ unsigned long rets;
+};
+
+/*
+ * Save stack-backtrace addresses into a stack_trace buffer.
+ */
+void save_stack_trace(struct stack_trace *trace)
+{
+ unsigned long low, high;
+ unsigned long fp;
+ struct stackframe *frame;
+ int skip = trace->skip;
+
+ low = (unsigned long)task_stack_page(current);
+ high = low + THREAD_SIZE;
+ fp = current_frame_pointer;
+
+ while (fp >= low && fp <= (high - sizeof(*frame))) {
+ frame = (struct stackframe *)fp;
+
+ if (skip) {
+ skip--;
+ } else {
+ trace->entries[trace->nr_entries++] = frame->rets;
+ if (trace->nr_entries >= trace->max_entries)
+ break;
+ }
+
+ /*
+ * The next frame must be at a higher address than the
+ * current frame.
+ */
+ low = fp + sizeof(*frame);
+ fp = frame->fp;
+ }
+}
+EXPORT_SYMBOL_GPL(save_stack_trace);
--
1.7.1
--
Sent by an employee of the Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum.
^ permalink raw reply related
* [PATCH v5 35/36] Hexagon: Add self to MAINTAINERS
From: Richard Kuo @ 2011-10-31 23:56 UTC (permalink / raw)
To: torvalds, linux-kernel
Signed-off-by: Richard Kuo <rkuo@codeaurora.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
---
MAINTAINERS | 6 ++++++
1 files changed, 6 insertions(+), 0 deletions(-)
diff --git a/MAINTAINERS b/MAINTAINERS
index ace8f9c..94a73fc 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -5288,6 +5288,12 @@ F: fs/qnx4/
F: include/linux/qnx4_fs.h
F: include/linux/qnxtypes.h
+QUALCOMM HEXAGON ARCHITECTURE
+M: Richard Kuo <rkuo@codeaurora.org>
+L: linux-hexagon@vger.kernel.org
+S: Supported
+F: arch/hexagon/
+
RADOS BLOCK DEVICE (RBD)
F: include/linux/qnxtypes.h
M: Yehuda Sadeh <yehuda@hq.newdream.net>
--
1.7.1
--
Sent by an employee of the Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum.
^ permalink raw reply related
* [PATCH v5 00/36] Hexagon: Add support for Qualcomm Hexagon architecture
From: Richard Kuo @ 2011-10-31 23:58 UTC (permalink / raw)
To: torvalds, linux-kernel
Linus,
This is the fifth version of the patchset (with one tiny whitespace fix) to
the Linux kernel to support the Qualcomm Hexagon architecture. I believe it
has all the acks necessary and is ready for inclusion in your tree.
Please let me know if there's anything else you require of these patches.
Thanks,
Richard Kuo
Richard Kuo (36):
Hexagon: Add generic headers
Hexagon: Core arch-specific header files
Hexagon: Add bitops support
Hexagon: Add atomic ops support
Hexagon: Add syscalls
Hexagon: Add processor and system headers
Hexagon: Add threadinfo
Hexagon: Add delay functions
Hexagon: Add checksum functions
Hexagon: Add memcpy and memset accelerated functions
Hexagon: Add hypervisor interface
Hexagon: Export ksyms defined in assembly files.
Hexagon: Support dynamic module loading.
Hexagon: Add signal functions
Hexagon: Add init_task and process functions
Hexagon: Add startup code
Hexagon: Add interrupts
Hexagon: Add time and timer functions
Hexagon: Add ptrace support
Hexagon: Provide basic debugging and system trap support.
Hexagon: Add SMP support
Hexagon: Add locking types and functions
Hexagon: Add user access functions
Hexagon: Provide basic implementation and/or stubs for I/O routines.
Hexagon: Implement basic cache-flush support
Hexagon: Implement basic TLB management routines for Hexagon.
Hexagon: Provide DMA implementation
Hexagon: Add ioremap support
Hexagon: Add page table header files & etc.
Hexagon: Add page-fault support.
Hexagon: kgdb support files
Hexagon: Comet platform support
Hexagon: Add configuration and makefiles for the Hexagon
architecture.
Hexagon: Add basic stacktrace functionality for Hexagon architecture.
Hexagon: Add self to MAINTAINERS
Add extra arch overrides to asm-generic/checksum.h
MAINTAINERS | 6 +
arch/hexagon/Kconfig | 220 ++++++++++++
arch/hexagon/Makefile | 59 +++
arch/hexagon/configs/comet_defconfig | 85 +++++
arch/hexagon/include/asm/Kbuild | 59 +++
arch/hexagon/include/asm/asm-offsets.h | 1 +
arch/hexagon/include/asm/atomic.h | 164 +++++++++
arch/hexagon/include/asm/bitops.h | 301 ++++++++++++++++
arch/hexagon/include/asm/bitsperlong.h | 26 ++
arch/hexagon/include/asm/byteorder.h | 28 ++
arch/hexagon/include/asm/cache.h | 34 ++
arch/hexagon/include/asm/cacheflush.h | 99 ++++++
arch/hexagon/include/asm/checksum.h | 50 +++
arch/hexagon/include/asm/delay.h | 28 ++
arch/hexagon/include/asm/dma-mapping.h | 101 ++++++
arch/hexagon/include/asm/dma.h | 29 ++
arch/hexagon/include/asm/elf.h | 229 ++++++++++++
arch/hexagon/include/asm/fixmap.h | 73 ++++
arch/hexagon/include/asm/fpu.h | 4 +
arch/hexagon/include/asm/futex.h | 137 ++++++++
arch/hexagon/include/asm/hexagon_vm.h | 281 +++++++++++++++
arch/hexagon/include/asm/intrinsics.h | 26 ++
arch/hexagon/include/asm/io.h | 326 +++++++++++++++++
arch/hexagon/include/asm/irq.h | 36 ++
arch/hexagon/include/asm/irqflags.h | 62 ++++
arch/hexagon/include/asm/kgdb.h | 43 +++
arch/hexagon/include/asm/linkage.h | 25 ++
arch/hexagon/include/asm/mem-layout.h | 112 ++++++
arch/hexagon/include/asm/mmu.h | 37 ++
arch/hexagon/include/asm/mmu_context.h | 100 ++++++
arch/hexagon/include/asm/module.h | 26 ++
arch/hexagon/include/asm/mutex.h | 8 +
arch/hexagon/include/asm/page.h | 157 +++++++++
arch/hexagon/include/asm/param.h | 26 ++
arch/hexagon/include/asm/perf_event.h | 24 ++
arch/hexagon/include/asm/pgalloc.h | 146 ++++++++
arch/hexagon/include/asm/pgtable.h | 518 +++++++++++++++++++++++++++
arch/hexagon/include/asm/processor.h | 123 +++++++
arch/hexagon/include/asm/ptrace.h | 35 ++
arch/hexagon/include/asm/registers.h | 236 +++++++++++++
arch/hexagon/include/asm/setup.h | 29 ++
arch/hexagon/include/asm/sigcontext.h | 33 ++
arch/hexagon/include/asm/signal.h | 26 ++
arch/hexagon/include/asm/smp.h | 44 +++
arch/hexagon/include/asm/spinlock.h | 186 ++++++++++
arch/hexagon/include/asm/spinlock_types.h | 42 +++
arch/hexagon/include/asm/string.h | 32 ++
arch/hexagon/include/asm/suspend.h | 27 ++
arch/hexagon/include/asm/swab.h | 24 ++
arch/hexagon/include/asm/syscall.h | 54 +++
arch/hexagon/include/asm/system.h | 126 +++++++
arch/hexagon/include/asm/thread_info.h | 154 ++++++++
arch/hexagon/include/asm/time.h | 29 ++
arch/hexagon/include/asm/timer-regs.h | 39 ++
arch/hexagon/include/asm/timex.h | 37 ++
arch/hexagon/include/asm/tlb.h | 39 ++
arch/hexagon/include/asm/tlbflush.h | 58 +++
arch/hexagon/include/asm/traps.h | 29 ++
arch/hexagon/include/asm/uaccess.h | 116 ++++++
arch/hexagon/include/asm/unistd.h | 36 ++
arch/hexagon/include/asm/user.h | 81 +++++
arch/hexagon/include/asm/vdso.h | 31 ++
arch/hexagon/include/asm/vm_fault.h | 26 ++
arch/hexagon/include/asm/vm_mmu.h | 111 ++++++
arch/hexagon/kernel/Makefile | 19 +
arch/hexagon/kernel/asm-offsets.c | 104 ++++++
arch/hexagon/kernel/dma.c | 220 ++++++++++++
arch/hexagon/kernel/head.S | 162 +++++++++
arch/hexagon/kernel/hexagon_ksyms.c | 42 +++
arch/hexagon/kernel/init_task.c | 54 +++
arch/hexagon/kernel/irq_cpu.c | 90 +++++
arch/hexagon/kernel/kgdb.c | 255 ++++++++++++++
arch/hexagon/kernel/module.c | 162 +++++++++
arch/hexagon/kernel/process.c | 279 +++++++++++++++
arch/hexagon/kernel/ptrace.c | 180 ++++++++++
arch/hexagon/kernel/reset.c | 38 ++
arch/hexagon/kernel/setup.c | 145 ++++++++
arch/hexagon/kernel/signal.c | 345 ++++++++++++++++++
arch/hexagon/kernel/smp.c | 276 +++++++++++++++
arch/hexagon/kernel/stacktrace.c | 66 ++++
arch/hexagon/kernel/syscall.c | 90 +++++
arch/hexagon/kernel/syscalltab.c | 32 ++
arch/hexagon/kernel/time.c | 250 +++++++++++++
arch/hexagon/kernel/topology.c | 52 +++
arch/hexagon/kernel/trampoline.S | 35 ++
arch/hexagon/kernel/traps.c | 454 ++++++++++++++++++++++++
arch/hexagon/kernel/vdso.c | 101 ++++++
arch/hexagon/kernel/vm_entry.S | 269 ++++++++++++++
arch/hexagon/kernel/vm_events.c | 101 ++++++
arch/hexagon/kernel/vm_init_segtable.S | 442 +++++++++++++++++++++++
arch/hexagon/kernel/vm_ops.S | 102 ++++++
arch/hexagon/kernel/vm_switch.S | 95 +++++
arch/hexagon/kernel/vm_vectors.S | 50 +++
arch/hexagon/kernel/vmlinux.lds.S | 93 +++++
arch/hexagon/lib/Makefile | 4 +
arch/hexagon/lib/checksum.c | 203 +++++++++++
arch/hexagon/lib/io.c | 91 +++++
arch/hexagon/lib/memcpy.S | 543 +++++++++++++++++++++++++++++
arch/hexagon/lib/memset.S | 316 +++++++++++++++++
arch/hexagon/mm/Makefile | 7 +
arch/hexagon/mm/cache.c | 128 +++++++
arch/hexagon/mm/copy_from_user.S | 114 ++++++
arch/hexagon/mm/copy_to_user.S | 92 +++++
arch/hexagon/mm/copy_user_template.S | 185 ++++++++++
arch/hexagon/mm/init.c | 276 +++++++++++++++
arch/hexagon/mm/ioremap.c | 56 +++
arch/hexagon/mm/pgalloc.c | 25 ++
arch/hexagon/mm/strnlen_user.S | 140 ++++++++
arch/hexagon/mm/uaccess.c | 59 +++
arch/hexagon/mm/vm_fault.c | 187 ++++++++++
arch/hexagon/mm/vm_tlb.c | 93 +++++
include/asm-generic/checksum.h | 4 +
include/asm-generic/rwsem.h | 132 +++++++
113 files changed, 12747 insertions(+), 0 deletions(-)
create mode 100644 arch/hexagon/Kconfig
create mode 100644 arch/hexagon/Makefile
create mode 100644 arch/hexagon/configs/comet_defconfig
create mode 100644 arch/hexagon/include/asm/Kbuild
create mode 100644 arch/hexagon/include/asm/asm-offsets.h
create mode 100644 arch/hexagon/include/asm/atomic.h
create mode 100644 arch/hexagon/include/asm/bitops.h
create mode 100644 arch/hexagon/include/asm/bitsperlong.h
create mode 100644 arch/hexagon/include/asm/byteorder.h
create mode 100644 arch/hexagon/include/asm/cache.h
create mode 100644 arch/hexagon/include/asm/cacheflush.h
create mode 100644 arch/hexagon/include/asm/checksum.h
create mode 100644 arch/hexagon/include/asm/delay.h
create mode 100644 arch/hexagon/include/asm/dma-mapping.h
create mode 100644 arch/hexagon/include/asm/dma.h
create mode 100644 arch/hexagon/include/asm/elf.h
create mode 100644 arch/hexagon/include/asm/fixmap.h
create mode 100644 arch/hexagon/include/asm/fpu.h
create mode 100644 arch/hexagon/include/asm/futex.h
create mode 100644 arch/hexagon/include/asm/hexagon_vm.h
create mode 100644 arch/hexagon/include/asm/intrinsics.h
create mode 100644 arch/hexagon/include/asm/io.h
create mode 100644 arch/hexagon/include/asm/irq.h
create mode 100644 arch/hexagon/include/asm/irqflags.h
create mode 100644 arch/hexagon/include/asm/kgdb.h
create mode 100644 arch/hexagon/include/asm/linkage.h
create mode 100644 arch/hexagon/include/asm/mem-layout.h
create mode 100644 arch/hexagon/include/asm/mmu.h
create mode 100644 arch/hexagon/include/asm/mmu_context.h
create mode 100644 arch/hexagon/include/asm/module.h
create mode 100644 arch/hexagon/include/asm/mutex.h
create mode 100644 arch/hexagon/include/asm/page.h
create mode 100644 arch/hexagon/include/asm/param.h
create mode 100644 arch/hexagon/include/asm/perf_event.h
create mode 100644 arch/hexagon/include/asm/pgalloc.h
create mode 100644 arch/hexagon/include/asm/pgtable.h
create mode 100644 arch/hexagon/include/asm/processor.h
create mode 100644 arch/hexagon/include/asm/ptrace.h
create mode 100644 arch/hexagon/include/asm/registers.h
create mode 100644 arch/hexagon/include/asm/setup.h
create mode 100644 arch/hexagon/include/asm/sigcontext.h
create mode 100644 arch/hexagon/include/asm/signal.h
create mode 100644 arch/hexagon/include/asm/smp.h
create mode 100644 arch/hexagon/include/asm/spinlock.h
create mode 100644 arch/hexagon/include/asm/spinlock_types.h
create mode 100644 arch/hexagon/include/asm/string.h
create mode 100644 arch/hexagon/include/asm/suspend.h
create mode 100644 arch/hexagon/include/asm/swab.h
create mode 100644 arch/hexagon/include/asm/syscall.h
create mode 100644 arch/hexagon/include/asm/system.h
create mode 100644 arch/hexagon/include/asm/thread_info.h
create mode 100644 arch/hexagon/include/asm/time.h
create mode 100644 arch/hexagon/include/asm/timer-regs.h
create mode 100644 arch/hexagon/include/asm/timex.h
create mode 100644 arch/hexagon/include/asm/tlb.h
create mode 100644 arch/hexagon/include/asm/tlbflush.h
create mode 100644 arch/hexagon/include/asm/traps.h
create mode 100644 arch/hexagon/include/asm/uaccess.h
create mode 100644 arch/hexagon/include/asm/unistd.h
create mode 100644 arch/hexagon/include/asm/user.h
create mode 100644 arch/hexagon/include/asm/vdso.h
create mode 100644 arch/hexagon/include/asm/vm_fault.h
create mode 100644 arch/hexagon/include/asm/vm_mmu.h
create mode 100644 arch/hexagon/kernel/Makefile
create mode 100644 arch/hexagon/kernel/asm-offsets.c
create mode 100644 arch/hexagon/kernel/dma.c
create mode 100644 arch/hexagon/kernel/head.S
create mode 100644 arch/hexagon/kernel/hexagon_ksyms.c
create mode 100644 arch/hexagon/kernel/init_task.c
create mode 100644 arch/hexagon/kernel/irq_cpu.c
create mode 100644 arch/hexagon/kernel/kgdb.c
create mode 100644 arch/hexagon/kernel/module.c
create mode 100644 arch/hexagon/kernel/process.c
create mode 100644 arch/hexagon/kernel/ptrace.c
create mode 100644 arch/hexagon/kernel/reset.c
create mode 100644 arch/hexagon/kernel/setup.c
create mode 100644 arch/hexagon/kernel/signal.c
create mode 100644 arch/hexagon/kernel/smp.c
create mode 100644 arch/hexagon/kernel/stacktrace.c
create mode 100644 arch/hexagon/kernel/syscall.c
create mode 100644 arch/hexagon/kernel/syscalltab.c
create mode 100644 arch/hexagon/kernel/time.c
create mode 100644 arch/hexagon/kernel/topology.c
create mode 100644 arch/hexagon/kernel/trampoline.S
create mode 100644 arch/hexagon/kernel/traps.c
create mode 100644 arch/hexagon/kernel/vdso.c
create mode 100644 arch/hexagon/kernel/vm_entry.S
create mode 100644 arch/hexagon/kernel/vm_events.c
create mode 100644 arch/hexagon/kernel/vm_init_segtable.S
create mode 100644 arch/hexagon/kernel/vm_ops.S
create mode 100644 arch/hexagon/kernel/vm_switch.S
create mode 100644 arch/hexagon/kernel/vm_vectors.S
create mode 100644 arch/hexagon/kernel/vmlinux.lds.S
create mode 100644 arch/hexagon/lib/Makefile
create mode 100644 arch/hexagon/lib/checksum.c
create mode 100644 arch/hexagon/lib/io.c
create mode 100644 arch/hexagon/lib/memcpy.S
create mode 100644 arch/hexagon/lib/memset.S
create mode 100644 arch/hexagon/mm/Makefile
create mode 100644 arch/hexagon/mm/cache.c
create mode 100644 arch/hexagon/mm/copy_from_user.S
create mode 100644 arch/hexagon/mm/copy_to_user.S
create mode 100644 arch/hexagon/mm/copy_user_template.S
create mode 100644 arch/hexagon/mm/init.c
create mode 100644 arch/hexagon/mm/ioremap.c
create mode 100644 arch/hexagon/mm/pgalloc.c
create mode 100644 arch/hexagon/mm/strnlen_user.S
create mode 100644 arch/hexagon/mm/uaccess.c
create mode 100644 arch/hexagon/mm/vm_fault.c
create mode 100644 arch/hexagon/mm/vm_tlb.c
create mode 100644 include/asm-generic/rwsem.h
--
Sent by an employee of the Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum.
^ permalink raw reply
* [PATCH v5 36/36] Add extra arch overrides to asm-generic/checksum.h
From: Richard Kuo @ 2011-10-31 23:56 UTC (permalink / raw)
To: torvalds, linux-kernel
There are plausible reasons for architectures to provide their own
versions of csum_partial_copy_nocheck and csum_tcpudp_magic.
By protecting these, the architecture can still re-use the
asm-generic checksum.h, instead of copying it.
Signed-off-by: Linas Vepstas <linas@codeaurora.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
---
include/asm-generic/checksum.h | 4 ++++
1 files changed, 4 insertions(+), 0 deletions(-)
diff --git a/include/asm-generic/checksum.h b/include/asm-generic/checksum.h
index 4647c76..c084767 100644
--- a/include/asm-generic/checksum.h
+++ b/include/asm-generic/checksum.h
@@ -33,8 +33,10 @@ extern __wsum csum_partial_copy(const void *src, void *dst, int len, __wsum sum)
extern __wsum csum_partial_copy_from_user(const void __user *src, void *dst,
int len, __wsum sum, int *csum_err);
+#ifndef csum_partial_copy_nocheck
#define csum_partial_copy_nocheck(src, dst, len, sum) \
csum_partial_copy((src), (dst), (len), (sum))
+#endif
/*
* This is a version of ip_compute_csum() optimized for IP headers,
@@ -63,12 +65,14 @@ csum_tcpudp_nofold(__be32 saddr, __be32 daddr, unsigned short len,
unsigned short proto, __wsum sum);
#endif
+#ifndef csum_tcpudp_magic
static inline __sum16
csum_tcpudp_magic(__be32 saddr, __be32 daddr, unsigned short len,
unsigned short proto, __wsum sum)
{
return csum_fold(csum_tcpudp_nofold(saddr, daddr, len, proto, sum));
}
+#endif
/*
* this routine is used for miscellaneous IP-like checksums, mainly
--
1.7.1
--
Sent by an employee of the Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum.
^ permalink raw reply related
* [PATCH v5 28/36] Hexagon: Add ioremap support
From: Richard Kuo @ 2011-10-31 23:52 UTC (permalink / raw)
To: torvalds, linux-kernel
Signed-off-by: Richard Kuo <rkuo@codeaurora.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
---
arch/hexagon/mm/ioremap.c | 56 +++++++++++++++++++++++++++++++++++++++++++++
1 files changed, 56 insertions(+), 0 deletions(-)
create mode 100644 arch/hexagon/mm/ioremap.c
diff --git a/arch/hexagon/mm/ioremap.c b/arch/hexagon/mm/ioremap.c
new file mode 100644
index 0000000..3a37bc3
--- /dev/null
+++ b/arch/hexagon/mm/ioremap.c
@@ -0,0 +1,56 @@
+/*
+ * I/O remap functions for Hexagon
+ *
+ * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+#include <linux/io.h>
+#include <linux/vmalloc.h>
+
+void __iomem *ioremap_nocache(unsigned long phys_addr, unsigned long size)
+{
+ unsigned long last_addr, addr;
+ unsigned long offset = phys_addr & ~PAGE_MASK;
+ struct vm_struct *area;
+
+ pgprot_t prot = __pgprot(_PAGE_PRESENT|_PAGE_READ|_PAGE_WRITE
+ |(__HEXAGON_C_DEV << 6));
+
+ last_addr = phys_addr + size - 1;
+
+ /* Wrapping not allowed */
+ if (!size || (last_addr < phys_addr))
+ return NULL;
+
+ /* Rounds up to next page size, including whole-page offset */
+ size = PAGE_ALIGN(offset + size);
+
+ area = get_vm_area(size, VM_IOREMAP);
+ addr = (unsigned long)area->addr;
+
+ if (ioremap_page_range(addr, addr+size, phys_addr, prot)) {
+ vunmap((void *)addr);
+ return NULL;
+ }
+
+ return (void __iomem *) (offset + addr);
+}
+
+void __iounmap(const volatile void __iomem *addr)
+{
+ vunmap((void *) ((unsigned long) addr & PAGE_MASK));
+}
--
1.7.1
--
Sent by an employee of the Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum.
^ permalink raw reply related
* [PATCH v5 31/36] Hexagon: kgdb support files
From: Richard Kuo @ 2011-10-31 23:54 UTC (permalink / raw)
To: torvalds, linux-kernel
Signed-off-by: Linas Vepstas <linas@codeaurora.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
---
arch/hexagon/include/asm/kgdb.h | 43 +++++++
arch/hexagon/kernel/kgdb.c | 255 +++++++++++++++++++++++++++++++++++++++
2 files changed, 298 insertions(+), 0 deletions(-)
create mode 100644 arch/hexagon/include/asm/kgdb.h
create mode 100644 arch/hexagon/kernel/kgdb.c
diff --git a/arch/hexagon/include/asm/kgdb.h b/arch/hexagon/include/asm/kgdb.h
new file mode 100644
index 0000000..9e87797
--- /dev/null
+++ b/arch/hexagon/include/asm/kgdb.h
@@ -0,0 +1,43 @@
+/*
+ * arch/hexagon/include/asm/kgdb.h - Hexagon KGDB Support
+ *
+ * Copyright (c) 2011, Code Aurora Forum. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+#ifndef __HEXAGON_KGDB_H__
+#define __HEXAGON_KGDB_H__
+
+#define BREAK_INSTR_SIZE 4
+#define CACHE_FLUSH_IS_SAFE 1
+#define BUFMAX ((NUMREGBYTES * 2) + 512)
+
+static inline void arch_kgdb_breakpoint(void)
+{
+ asm("trap0(#0xDB)");
+}
+
+/* Registers:
+ * 32 gpr + sa0/1 + lc0/1 + m0/1 + gp + ugp + pred + pc = 42 total.
+ * vm regs = psp+elr+est+badva = 4
+ * syscall+restart = 2 more
+ * so 48 = 42 +4 + 2
+ */
+#define DBG_USER_REGS 42
+#define DBG_MAX_REG_NUM (DBG_USER_REGS + 6)
+#define NUMREGBYTES (DBG_MAX_REG_NUM*4)
+
+#endif /* __HEXAGON_KGDB_H__ */
diff --git a/arch/hexagon/kernel/kgdb.c b/arch/hexagon/kernel/kgdb.c
new file mode 100644
index 0000000..38ab0ea
--- /dev/null
+++ b/arch/hexagon/kernel/kgdb.c
@@ -0,0 +1,255 @@
+/*
+ * arch/hexagon/kernel/kgdb.c - Hexagon KGDB Support
+ *
+ * Copyright (c) 2011, Code Aurora Forum. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+#include <linux/kdebug.h>
+#include <linux/kgdb.h>
+
+/* All registers are 4 bytes, for now */
+#define GDB_SIZEOF_REG 4
+
+/* The register names are used during printing of the regs;
+ * Keep these at three letters to pretty-print. */
+struct dbg_reg_def_t dbg_reg_def[DBG_MAX_REG_NUM] = {
+ { " r0", GDB_SIZEOF_REG, offsetof(struct pt_regs, r00)},
+ { " r1", GDB_SIZEOF_REG, offsetof(struct pt_regs, r01)},
+ { " r2", GDB_SIZEOF_REG, offsetof(struct pt_regs, r02)},
+ { " r3", GDB_SIZEOF_REG, offsetof(struct pt_regs, r03)},
+ { " r4", GDB_SIZEOF_REG, offsetof(struct pt_regs, r04)},
+ { " r5", GDB_SIZEOF_REG, offsetof(struct pt_regs, r05)},
+ { " r6", GDB_SIZEOF_REG, offsetof(struct pt_regs, r06)},
+ { " r7", GDB_SIZEOF_REG, offsetof(struct pt_regs, r07)},
+ { " r8", GDB_SIZEOF_REG, offsetof(struct pt_regs, r08)},
+ { " r9", GDB_SIZEOF_REG, offsetof(struct pt_regs, r09)},
+ { "r10", GDB_SIZEOF_REG, offsetof(struct pt_regs, r10)},
+ { "r11", GDB_SIZEOF_REG, offsetof(struct pt_regs, r11)},
+ { "r12", GDB_SIZEOF_REG, offsetof(struct pt_regs, r12)},
+ { "r13", GDB_SIZEOF_REG, offsetof(struct pt_regs, r13)},
+ { "r14", GDB_SIZEOF_REG, offsetof(struct pt_regs, r14)},
+ { "r15", GDB_SIZEOF_REG, offsetof(struct pt_regs, r15)},
+ { "r16", GDB_SIZEOF_REG, offsetof(struct pt_regs, r16)},
+ { "r17", GDB_SIZEOF_REG, offsetof(struct pt_regs, r17)},
+ { "r18", GDB_SIZEOF_REG, offsetof(struct pt_regs, r18)},
+ { "r19", GDB_SIZEOF_REG, offsetof(struct pt_regs, r19)},
+ { "r20", GDB_SIZEOF_REG, offsetof(struct pt_regs, r20)},
+ { "r21", GDB_SIZEOF_REG, offsetof(struct pt_regs, r21)},
+ { "r22", GDB_SIZEOF_REG, offsetof(struct pt_regs, r22)},
+ { "r23", GDB_SIZEOF_REG, offsetof(struct pt_regs, r23)},
+ { "r24", GDB_SIZEOF_REG, offsetof(struct pt_regs, r24)},
+ { "r25", GDB_SIZEOF_REG, offsetof(struct pt_regs, r25)},
+ { "r26", GDB_SIZEOF_REG, offsetof(struct pt_regs, r26)},
+ { "r27", GDB_SIZEOF_REG, offsetof(struct pt_regs, r27)},
+ { "r28", GDB_SIZEOF_REG, offsetof(struct pt_regs, r28)},
+ { "r29", GDB_SIZEOF_REG, offsetof(struct pt_regs, r29)},
+ { "r30", GDB_SIZEOF_REG, offsetof(struct pt_regs, r30)},
+ { "r31", GDB_SIZEOF_REG, offsetof(struct pt_regs, r31)},
+
+ { "usr", GDB_SIZEOF_REG, offsetof(struct pt_regs, usr)},
+ { "preds", GDB_SIZEOF_REG, offsetof(struct pt_regs, preds)},
+ { " m0", GDB_SIZEOF_REG, offsetof(struct pt_regs, m0)},
+ { " m1", GDB_SIZEOF_REG, offsetof(struct pt_regs, m1)},
+ { "sa0", GDB_SIZEOF_REG, offsetof(struct pt_regs, sa0)},
+ { "sa1", GDB_SIZEOF_REG, offsetof(struct pt_regs, sa1)},
+ { "lc0", GDB_SIZEOF_REG, offsetof(struct pt_regs, lc0)},
+ { "lc1", GDB_SIZEOF_REG, offsetof(struct pt_regs, lc1)},
+ { " gp", GDB_SIZEOF_REG, offsetof(struct pt_regs, gp)},
+ { "ugp", GDB_SIZEOF_REG, offsetof(struct pt_regs, ugp)},
+ { "psp", GDB_SIZEOF_REG, offsetof(struct pt_regs, hvmer.vmpsp)},
+ { "elr", GDB_SIZEOF_REG, offsetof(struct pt_regs, hvmer.vmel)},
+ { "est", GDB_SIZEOF_REG, offsetof(struct pt_regs, hvmer.vmest)},
+ { "badva", GDB_SIZEOF_REG, offsetof(struct pt_regs, hvmer.vmbadva)},
+ { "restart_r0", GDB_SIZEOF_REG, offsetof(struct pt_regs, restart_r0)},
+ { "syscall_nr", GDB_SIZEOF_REG, offsetof(struct pt_regs, syscall_nr)},
+};
+
+struct kgdb_arch arch_kgdb_ops = {
+ /* trap0(#0xDB) 0x0cdb0054 */
+ .gdb_bpt_instr = {0x54, 0x00, 0xdb, 0x0c},
+};
+
+char *dbg_get_reg(int regno, void *mem, struct pt_regs *regs)
+{
+ if (regno >= DBG_MAX_REG_NUM || regno < 0)
+ return NULL;
+
+ *((unsigned long *) mem) = *((unsigned long *) ((void *)regs +
+ dbg_reg_def[regno].offset));
+
+ return dbg_reg_def[regno].name;
+}
+
+int dbg_set_reg(int regno, void *mem, struct pt_regs *regs)
+{
+ if (regno >= DBG_MAX_REG_NUM || regno < 0)
+ return -EINVAL;
+
+ *((unsigned long *) ((void *)regs + dbg_reg_def[regno].offset)) =
+ *((unsigned long *) mem);
+
+ return 0;
+}
+
+void kgdb_arch_set_pc(struct pt_regs *regs, unsigned long pc)
+{
+ instruction_pointer(regs) = pc;
+}
+
+#ifdef CONFIG_SMP
+
+/**
+ * kgdb_roundup_cpus - Get other CPUs into a holding pattern
+ * @flags: Current IRQ state
+ *
+ * On SMP systems, we need to get the attention of the other CPUs
+ * and get them be in a known state. This should do what is needed
+ * to get the other CPUs to call kgdb_wait(). Note that on some arches,
+ * the NMI approach is not used for rounding up all the CPUs. For example,
+ * in case of MIPS, smp_call_function() is used to roundup CPUs. In
+ * this case, we have to make sure that interrupts are enabled before
+ * calling smp_call_function(). The argument to this function is
+ * the flags that will be used when restoring the interrupts. There is
+ * local_irq_save() call before kgdb_roundup_cpus().
+ *
+ * On non-SMP systems, this is not called.
+ */
+
+static void hexagon_kgdb_nmi_hook(void *ignored)
+{
+ kgdb_nmicallback(raw_smp_processor_id(), get_irq_regs());
+}
+
+void kgdb_roundup_cpus(unsigned long flags)
+{
+ local_irq_enable();
+ smp_call_function(hexagon_kgdb_nmi_hook, NULL, 0);
+ local_irq_disable();
+}
+#endif
+
+
+/* Not yet working */
+void sleeping_thread_to_gdb_regs(unsigned long *gdb_regs,
+ struct task_struct *task)
+{
+ struct pt_regs *thread_regs;
+
+ if (task == NULL)
+ return;
+
+ /* Initialize to zero */
+ memset(gdb_regs, 0, NUMREGBYTES);
+
+ /* Otherwise, we have only some registers from switch_to() */
+ thread_regs = task_pt_regs(task);
+ gdb_regs[0] = thread_regs->r00;
+}
+
+/**
+ * kgdb_arch_handle_exception - Handle architecture specific GDB packets.
+ * @vector: The error vector of the exception that happened.
+ * @signo: The signal number of the exception that happened.
+ * @err_code: The error code of the exception that happened.
+ * @remcom_in_buffer: The buffer of the packet we have read.
+ * @remcom_out_buffer: The buffer of %BUFMAX bytes to write a packet into.
+ * @regs: The &struct pt_regs of the current process.
+ *
+ * This function MUST handle the 'c' and 's' command packets,
+ * as well packets to set / remove a hardware breakpoint, if used.
+ * If there are additional packets which the hardware needs to handle,
+ * they are handled here. The code should return -1 if it wants to
+ * process more packets, and a %0 or %1 if it wants to exit from the
+ * kgdb callback.
+ *
+ * Not yet working.
+ */
+int kgdb_arch_handle_exception(int vector, int signo, int err_code,
+ char *remcom_in_buffer, char *remcom_out_buffer,
+ struct pt_regs *linux_regs)
+{
+ switch (remcom_in_buffer[0]) {
+ case 's':
+ case 'c':
+ return 0;
+ }
+ /* Stay in the debugger. */
+ return -1;
+}
+
+static int __kgdb_notify(struct die_args *args, unsigned long cmd)
+{
+ /* cpu roundup */
+ if (atomic_read(&kgdb_active) != -1) {
+ kgdb_nmicallback(smp_processor_id(), args->regs);
+ return NOTIFY_STOP;
+ }
+
+ if (user_mode(args->regs))
+ return NOTIFY_DONE;
+
+ if (kgdb_handle_exception(args->trapnr & 0xff, args->signr, args->err,
+ args->regs))
+ return NOTIFY_DONE;
+
+ return NOTIFY_STOP;
+}
+
+static int
+kgdb_notify(struct notifier_block *self, unsigned long cmd, void *ptr)
+{
+ unsigned long flags;
+ int ret;
+
+ local_irq_save(flags);
+ ret = __kgdb_notify(ptr, cmd);
+ local_irq_restore(flags);
+
+ return ret;
+}
+
+static struct notifier_block kgdb_notifier = {
+ .notifier_call = kgdb_notify,
+
+ /*
+ * Lowest-prio notifier priority, we want to be notified last:
+ */
+ .priority = -INT_MAX,
+};
+
+/**
+ * kgdb_arch_init - Perform any architecture specific initalization.
+ *
+ * This function will handle the initalization of any architecture
+ * specific callbacks.
+ */
+int kgdb_arch_init(void)
+{
+ return register_die_notifier(&kgdb_notifier);
+}
+
+/**
+ * kgdb_arch_exit - Perform any architecture specific uninitalization.
+ *
+ * This function will handle the uninitalization of any architecture
+ * specific callbacks, for dynamic registration and unregistration.
+ */
+void kgdb_arch_exit(void)
+{
+ unregister_die_notifier(&kgdb_notifier);
+}
+
--
1.7.1
--
Sent by an employee of the Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum.
^ permalink raw reply related
* [PATCH v5 26/36] Hexagon: Implement basic TLB management routines for Hexagon.
From: Richard Kuo @ 2011-10-31 23:52 UTC (permalink / raw)
To: torvalds, linux-kernel
Mostly all stubs, as the TLB is managed by the hypervisor.
Signed-off-by: Richard Kuo <rkuo@codeaurora.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
---
arch/hexagon/include/asm/tlb.h | 39 +++++++++++++++
arch/hexagon/include/asm/tlbflush.h | 58 ++++++++++++++++++++++
arch/hexagon/mm/vm_tlb.c | 93 +++++++++++++++++++++++++++++++++++
3 files changed, 190 insertions(+), 0 deletions(-)
create mode 100644 arch/hexagon/include/asm/tlb.h
create mode 100644 arch/hexagon/include/asm/tlbflush.h
create mode 100644 arch/hexagon/mm/vm_tlb.c
diff --git a/arch/hexagon/include/asm/tlb.h b/arch/hexagon/include/asm/tlb.h
new file mode 100644
index 0000000..473abde
--- /dev/null
+++ b/arch/hexagon/include/asm/tlb.h
@@ -0,0 +1,39 @@
+/*
+ * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+#ifndef _ASM_TLB_H
+#define _ASM_TLB_H
+
+#include <linux/pagemap.h>
+#include <asm/tlbflush.h>
+
+/*
+ * We don't need any special per-pte or per-vma handling...
+ */
+#define tlb_start_vma(tlb, vma) do { } while (0)
+#define tlb_end_vma(tlb, vma) do { } while (0)
+#define __tlb_remove_tlb_entry(tlb, ptep, address) do { } while (0)
+
+/*
+ * .. because we flush the whole mm when it fills up
+ */
+#define tlb_flush(tlb) flush_tlb_mm((tlb)->mm)
+
+#include <asm-generic/tlb.h>
+
+#endif
diff --git a/arch/hexagon/include/asm/tlbflush.h b/arch/hexagon/include/asm/tlbflush.h
new file mode 100644
index 0000000..b89a902
--- /dev/null
+++ b/arch/hexagon/include/asm/tlbflush.h
@@ -0,0 +1,58 @@
+/*
+ * TLB flush support for Hexagon
+ *
+ * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+#ifndef _ASM_TLBFLUSH_H
+#define _ASM_TLBFLUSH_H
+
+#include <linux/mm.h>
+#include <asm/processor.h>
+
+/*
+ * TLB flushing -- in "SMP", these routines get defined to be the
+ * ones from smp.c, else they are some local flavors.
+ */
+
+/*
+ * These functions are commonly macros, but in the interests of
+ * VM vs. native implementation and code size, we simply declare
+ * the function prototypes here.
+ */
+extern void tlb_flush_all(void);
+extern void flush_tlb_mm(struct mm_struct *mm);
+extern void flush_tlb_page(struct vm_area_struct *vma, unsigned long addr);
+extern void flush_tlb_range(struct vm_area_struct *vma,
+ unsigned long start, unsigned long end);
+extern void flush_tlb_kernel_range(unsigned long start, unsigned long end);
+extern void flush_tlb_one(unsigned long);
+
+/*
+ * "This is called in munmap when we have freed up some page-table pages.
+ * We don't need to do anything here..."
+ *
+ * The VM kernel doesn't walk page tables, and they are passed to the VMM
+ * by logical address. There doesn't seem to be any possibility that they
+ * could be referenced by the VM kernel based on a stale mapping, since
+ * they would only be located by consulting the mm structure, and they
+ * will have been purged from that structure by the munmap. Seems like
+ * a noop on HVM as well.
+ */
+#define flush_tlb_pgtables(mm, start, end)
+
+#endif
diff --git a/arch/hexagon/mm/vm_tlb.c b/arch/hexagon/mm/vm_tlb.c
new file mode 100644
index 0000000..c6ff415
--- /dev/null
+++ b/arch/hexagon/mm/vm_tlb.c
@@ -0,0 +1,93 @@
+/*
+ * Hexagon Virtual Machine TLB functions
+ *
+ * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+/*
+ * The Hexagon Virtual Machine conceals the real workings of
+ * the TLB, but there are one or two functions that need to
+ * be instantiated for it, differently from a native build.
+ */
+#include <linux/mm.h>
+#include <asm/page.h>
+#include <asm/hexagon_vm.h>
+
+/*
+ * Initial VM implementation has only one map active at a time, with
+ * TLB purgings on changes. So either we're nuking the current map,
+ * or it's a no-op. This operation is messy on true SMPs where other
+ * processors must be induced to flush the copies in their local TLBs,
+ * but Hexagon thread-based virtual processors share the same MMU.
+ */
+void flush_tlb_range(struct vm_area_struct *vma, unsigned long start,
+ unsigned long end)
+{
+ struct mm_struct *mm = vma->vm_mm;
+
+ if (mm->context.ptbase == current->active_mm->context.ptbase)
+ __vmclrmap((void *)start, end - start);
+}
+
+/*
+ * Flush a page from the kernel virtual map - used by highmem
+ */
+void flush_tlb_one(unsigned long vaddr)
+{
+ __vmclrmap((void *)vaddr, PAGE_SIZE);
+}
+
+/*
+ * Flush all TLBs across all CPUs, virtual or real.
+ * A single Hexagon core has 6 thread contexts but
+ * only one TLB.
+ */
+void tlb_flush_all(void)
+{
+ /* should probably use that fixaddr end or whateve label */
+ __vmclrmap(0, 0xffff0000);
+}
+
+/*
+ * Flush TLB entries associated with a given mm_struct mapping.
+ */
+void flush_tlb_mm(struct mm_struct *mm)
+{
+ /* Current Virtual Machine has only one map active at a time */
+ if (current->active_mm->context.ptbase == mm->context.ptbase)
+ tlb_flush_all();
+}
+
+/*
+ * Flush TLB state associated with a page of a vma.
+ */
+void flush_tlb_page(struct vm_area_struct *vma, unsigned long vaddr)
+{
+ struct mm_struct *mm = vma->vm_mm;
+
+ if (mm->context.ptbase == current->active_mm->context.ptbase)
+ __vmclrmap((void *)vaddr, PAGE_SIZE);
+}
+
+/*
+ * Flush TLB entries associated with a kernel address range.
+ * Like flush range, but without the check on the vma->vm_mm.
+ */
+void flush_tlb_kernel_range(unsigned long start, unsigned long end)
+{
+ __vmclrmap((void *)start, end - start);
+}
--
1.7.1
--
Sent by an employee of the Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum.
^ permalink raw reply related
* [PATCH v5 23/36] Hexagon: Add user access functions
From: Richard Kuo @ 2011-10-31 23:48 UTC (permalink / raw)
To: torvalds, linux-kernel
Signed-off-by: Richard Kuo <rkuo@codeaurora.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
---
arch/hexagon/include/asm/uaccess.h | 116 +++++++++++++++++++++
arch/hexagon/mm/copy_from_user.S | 114 +++++++++++++++++++++
arch/hexagon/mm/copy_to_user.S | 92 +++++++++++++++++
arch/hexagon/mm/copy_user_template.S | 185 ++++++++++++++++++++++++++++++++++
arch/hexagon/mm/strnlen_user.S | 140 +++++++++++++++++++++++++
arch/hexagon/mm/uaccess.c | 59 +++++++++++
6 files changed, 706 insertions(+), 0 deletions(-)
create mode 100644 arch/hexagon/include/asm/uaccess.h
create mode 100644 arch/hexagon/mm/copy_from_user.S
create mode 100644 arch/hexagon/mm/copy_to_user.S
create mode 100644 arch/hexagon/mm/copy_user_template.S
create mode 100644 arch/hexagon/mm/strnlen_user.S
create mode 100644 arch/hexagon/mm/uaccess.c
diff --git a/arch/hexagon/include/asm/uaccess.h b/arch/hexagon/include/asm/uaccess.h
new file mode 100644
index 0000000..7e706ea
--- /dev/null
+++ b/arch/hexagon/include/asm/uaccess.h
@@ -0,0 +1,116 @@
+/*
+ * User memory access support for Hexagon
+ *
+ * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+#ifndef _ASM_UACCESS_H
+#define _ASM_UACCESS_H
+/*
+ * User space memory access functions
+ */
+#include <linux/sched.h>
+#include <linux/mm.h>
+#include <asm/segment.h>
+#include <asm/sections.h>
+
+/*
+ * access_ok: - Checks if a user space pointer is valid
+ * @type: Type of access: %VERIFY_READ or %VERIFY_WRITE. Note that
+ * %VERIFY_WRITE is a superset of %VERIFY_READ - if it is safe
+ * to write to a block, it is always safe to read from it.
+ * @addr: User space pointer to start of block to check
+ * @size: Size of block to check
+ *
+ * Context: User context only. This function may sleep.
+ *
+ * Checks if a pointer to a block of memory in user space is valid.
+ *
+ * Returns true (nonzero) if the memory block *may* be valid, false (zero)
+ * if it is definitely invalid.
+ *
+ * User address space in Hexagon, like x86, goes to 0xbfffffff, so the
+ * simple MSB-based tests used by MIPS won't work. Some further
+ * optimization is probably possible here, but for now, keep it
+ * reasonably simple and not *too* slow. After all, we've got the
+ * MMU for backup.
+ */
+#define VERIFY_READ 0
+#define VERIFY_WRITE 1
+
+#define __access_ok(addr, size) \
+ ((get_fs().seg == KERNEL_DS.seg) || \
+ (((unsigned long)addr < get_fs().seg) && \
+ (unsigned long)size < (get_fs().seg - (unsigned long)addr)))
+
+/*
+ * When a kernel-mode page fault is taken, the faulting instruction
+ * address is checked against a table of exception_table_entries.
+ * Each entry is a tuple of the address of an instruction that may
+ * be authorized to fault, and the address at which execution should
+ * be resumed instead of the faulting instruction, so as to effect
+ * a workaround.
+ */
+
+/* Assembly somewhat optimized copy routines */
+unsigned long __copy_from_user_hexagon(void *to, const void __user *from,
+ unsigned long n);
+unsigned long __copy_to_user_hexagon(void __user *to, const void *from,
+ unsigned long n);
+
+#define __copy_from_user(to, from, n) __copy_from_user_hexagon(to, from, n)
+#define __copy_to_user(to, from, n) __copy_to_user_hexagon(to, from, n)
+
+/*
+ * XXX todo: some additonal performance gain is possible by
+ * implementing __copy_to/from_user_inatomic, which is much
+ * like __copy_to/from_user, but performs slightly less checking.
+ */
+
+__kernel_size_t __clear_user_hexagon(void __user *dest, unsigned long count);
+#define __clear_user(a, s) __clear_user_hexagon((a), (s))
+
+#define __strncpy_from_user(dst, src, n) hexagon_strncpy_from_user(dst, src, n)
+
+/* get around the ifndef in asm-generic/uaccess.h */
+#define __strnlen_user __strnlen_user
+
+extern long __strnlen_user(const char __user *src, long n);
+
+static inline long hexagon_strncpy_from_user(char *dst, const char __user *src,
+ long n);
+
+#include <asm-generic/uaccess.h>
+
+/* Todo: an actual accelerated version of this. */
+static inline long hexagon_strncpy_from_user(char *dst, const char __user *src,
+ long n)
+{
+ long res = __strnlen_user(src, n);
+
+ /* return from strnlen can't be zero -- that would be rubbish. */
+
+ if (res > n) {
+ copy_from_user(dst, src, n);
+ return n;
+ } else {
+ copy_from_user(dst, src, res);
+ return res-1;
+ }
+}
+
+#endif
diff --git a/arch/hexagon/mm/copy_from_user.S b/arch/hexagon/mm/copy_from_user.S
new file mode 100644
index 0000000..8eb1d4d
--- /dev/null
+++ b/arch/hexagon/mm/copy_from_user.S
@@ -0,0 +1,114 @@
+/*
+ * User memory copy functions for kernel
+ *
+ * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+/*
+ * The right way to do this involves valignb
+ * The easy way to do this is only speed up src/dest similar alignment.
+ */
+
+/*
+ * Copy to/from user are the same, except that for packets with a load and
+ * a store, I don't know how to tell which kind of exception we got.
+ * Therefore, we duplicate the function, and handle faulting addresses
+ * differently for each function
+ */
+
+/*
+ * copy from user: loads can fault
+ */
+#define src_sav r13
+#define dst_sav r12
+#define src_dst_sav r13:12
+#define d_dbuf r15:14
+#define w_dbuf r15
+
+#define dst r0
+#define src r1
+#define bytes r2
+#define loopcount r5
+
+#define FUNCNAME __copy_from_user_hexagon
+#include "copy_user_template.S"
+
+ /* LOAD FAULTS from COPY_FROM_USER */
+
+ /* Alignment loop. r2 has been updated. Return it. */
+ .falign
+1009:
+2009:
+4009:
+ {
+ r0 = r2
+ jumpr r31
+ }
+ /* Normal copy loops. Do epilog. Use src-src_sav to compute distance */
+ /* X - (A - B) == X + B - A */
+ .falign
+8089:
+ {
+ memd(dst) = d_dbuf
+ r2 += sub(src_sav,src)
+ }
+ {
+ r0 = r2
+ jumpr r31
+ }
+ .falign
+4089:
+ {
+ memw(dst) = w_dbuf
+ r2 += sub(src_sav,src)
+ }
+ {
+ r0 = r2
+ jumpr r31
+ }
+ .falign
+2089:
+ {
+ memh(dst) = w_dbuf
+ r2 += sub(src_sav,src)
+ }
+ {
+ r0 = r2
+ jumpr r31
+ }
+ .falign
+1089:
+ {
+ memb(dst) = w_dbuf
+ r2 += sub(src_sav,src)
+ }
+ {
+ r0 = r2
+ jumpr r31
+ }
+
+ /* COPY FROM USER: only loads can fail */
+
+ .section __ex_table,"a"
+ .long 1000b,1009b
+ .long 2000b,2009b
+ .long 4000b,4009b
+ .long 8080b,8089b
+ .long 4080b,4089b
+ .long 2080b,2089b
+ .long 1080b,1089b
+ .previous
diff --git a/arch/hexagon/mm/copy_to_user.S b/arch/hexagon/mm/copy_to_user.S
new file mode 100644
index 0000000..cb9740e
--- /dev/null
+++ b/arch/hexagon/mm/copy_to_user.S
@@ -0,0 +1,92 @@
+/*
+ * User memory copying routines for the Hexagon Kernel
+ *
+ * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+/* The right way to do this involves valignb
+ * The easy way to do this is only speed up src/dest similar alignment.
+ */
+
+/*
+ * Copy to/from user are the same, except that for packets with a load and
+ * a store, I don't know how to tell which kind of exception we got.
+ * Therefore, we duplicate the function, and handle faulting addresses
+ * differently for each function
+ */
+
+/*
+ * copy to user: stores can fault
+ */
+#define src_sav r13
+#define dst_sav r12
+#define src_dst_sav r13:12
+#define d_dbuf r15:14
+#define w_dbuf r15
+
+#define dst r0
+#define src r1
+#define bytes r2
+#define loopcount r5
+
+#define FUNCNAME __copy_to_user_hexagon
+#include "copy_user_template.S"
+
+ /* STORE FAULTS from COPY_TO_USER */
+ .falign
+1109:
+2109:
+4109:
+ /* Alignment loop. r2 has been updated. Return it. */
+ {
+ r0 = r2
+ jumpr r31
+ }
+ /* Normal copy loops. Use dst-dst_sav to compute distance */
+ /* dst holds best write, no need to unwind any loops */
+ /* X - (A - B) == X + B - A */
+ .falign
+8189:
+8199:
+4189:
+4199:
+2189:
+2199:
+1189:
+1199:
+ {
+ r2 += sub(dst_sav,dst)
+ }
+ {
+ r0 = r2
+ jumpr r31
+ }
+
+ /* COPY TO USER: only stores can fail */
+ .section __ex_table,"a"
+ .long 1100b,1109b
+ .long 2100b,2109b
+ .long 4100b,4109b
+ .long 8180b,8189b
+ .long 8190b,8199b
+ .long 4180b,4189b
+ .long 4190b,4199b
+ .long 2180b,2189b
+ .long 2190b,2199b
+ .long 1180b,1189b
+ .long 1190b,1199b
+ .previous
diff --git a/arch/hexagon/mm/copy_user_template.S b/arch/hexagon/mm/copy_user_template.S
new file mode 100644
index 0000000..08d7d7b
--- /dev/null
+++ b/arch/hexagon/mm/copy_user_template.S
@@ -0,0 +1,185 @@
+/*
+ * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+/* Numerology:
+ * WXYZ
+ * W: width in bytes
+ * X: Load=0, Store=1
+ * Y: Location 0=preamble,8=loop,9=epilog
+ * Z: Location=0,handler=9
+ */
+ .text
+ .global FUNCNAME
+ .type FUNCNAME, @function
+ .p2align 5
+FUNCNAME:
+ {
+ p0 = cmp.gtu(bytes,#0)
+ if (!p0.new) jump:nt .Ldone
+ r3 = or(dst,src)
+ r4 = xor(dst,src)
+ }
+ {
+ p1 = cmp.gtu(bytes,#15)
+ p0 = bitsclr(r3,#7)
+ if (!p0.new) jump:nt .Loop_not_aligned_8
+ src_dst_sav = combine(src,dst)
+ }
+
+ {
+ loopcount = lsr(bytes,#3)
+ if (!p1) jump .Lsmall
+ }
+ p3=sp1loop0(.Loop8,loopcount)
+.Loop8:
+8080:
+8180:
+ {
+ if (p3) memd(dst++#8) = d_dbuf
+ d_dbuf = memd(src++#8)
+ }:endloop0
+8190:
+ {
+ memd(dst++#8) = d_dbuf
+ bytes -= asl(loopcount,#3)
+ jump .Lsmall
+ }
+
+.Loop_not_aligned_8:
+ {
+ p0 = bitsclr(r4,#7)
+ if (p0.new) jump:nt .Lalign
+ }
+ {
+ p0 = bitsclr(r3,#3)
+ if (!p0.new) jump:nt .Loop_not_aligned_4
+ p1 = cmp.gtu(bytes,#7)
+ }
+
+ {
+ if (!p1) jump .Lsmall
+ loopcount = lsr(bytes,#2)
+ }
+ p3=sp1loop0(.Loop4,loopcount)
+.Loop4:
+4080:
+4180:
+ {
+ if (p3) memw(dst++#4) = w_dbuf
+ w_dbuf = memw(src++#4)
+ }:endloop0
+4190:
+ {
+ memw(dst++#4) = w_dbuf
+ bytes -= asl(loopcount,#2)
+ jump .Lsmall
+ }
+
+.Loop_not_aligned_4:
+ {
+ p0 = bitsclr(r3,#1)
+ if (!p0.new) jump:nt .Loop_not_aligned
+ p1 = cmp.gtu(bytes,#3)
+ }
+
+ {
+ if (!p1) jump .Lsmall
+ loopcount = lsr(bytes,#1)
+ }
+ p3=sp1loop0(.Loop2,loopcount)
+.Loop2:
+2080:
+2180:
+ {
+ if (p3) memh(dst++#2) = w_dbuf
+ w_dbuf = memuh(src++#2)
+ }:endloop0
+2190:
+ {
+ memh(dst++#2) = w_dbuf
+ bytes -= asl(loopcount,#1)
+ jump .Lsmall
+ }
+
+.Loop_not_aligned: /* Works for as small as one byte */
+ p3=sp1loop0(.Loop1,bytes)
+.Loop1:
+1080:
+1180:
+ {
+ if (p3) memb(dst++#1) = w_dbuf
+ w_dbuf = memub(src++#1)
+ }:endloop0
+ /* Done */
+1190:
+ {
+ memb(dst) = w_dbuf
+ jumpr r31
+ r0 = #0
+ }
+
+.Lsmall:
+ {
+ p0 = cmp.gtu(bytes,#0)
+ if (p0.new) jump:nt .Loop_not_aligned
+ }
+.Ldone:
+ {
+ r0 = #0
+ jumpr r31
+ }
+ .falign
+.Lalign:
+1000:
+ {
+ if (p0.new) w_dbuf = memub(src)
+ p0 = tstbit(src,#0)
+ if (!p1) jump .Lsmall
+ }
+1100:
+ {
+ if (p0) memb(dst++#1) = w_dbuf
+ if (p0) bytes = add(bytes,#-1)
+ if (p0) src = add(src,#1)
+ }
+2000:
+ {
+ if (p0.new) w_dbuf = memuh(src)
+ p0 = tstbit(src,#1)
+ if (!p1) jump .Lsmall
+ }
+2100:
+ {
+ if (p0) memh(dst++#2) = w_dbuf
+ if (p0) bytes = add(bytes,#-2)
+ if (p0) src = add(src,#2)
+ }
+4000:
+ {
+ if (p0.new) w_dbuf = memw(src)
+ p0 = tstbit(src,#2)
+ if (!p1) jump .Lsmall
+ }
+4100:
+ {
+ if (p0) memw(dst++#4) = w_dbuf
+ if (p0) bytes = add(bytes,#-4)
+ if (p0) src = add(src,#4)
+ jump FUNCNAME
+ }
+ .size FUNCNAME,.-FUNCNAME
diff --git a/arch/hexagon/mm/strnlen_user.S b/arch/hexagon/mm/strnlen_user.S
new file mode 100644
index 0000000..324ff54
--- /dev/null
+++ b/arch/hexagon/mm/strnlen_user.S
@@ -0,0 +1,140 @@
+/*
+ * User string length functions for kernel
+ *
+ * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+#define isrc r0
+#define max r1 /* Do not change! */
+
+#define end r2
+#define tmp1 r3
+
+#define obo r6 /* off-by-one */
+#define start r7
+#define mod8 r8
+#define dbuf r15:14
+#define dcmp r13:12
+
+/*
+ * The vector mask version of this turned out *really* badly.
+ * The hardware loop version also turned out *really* badly.
+ * Seems straight pointer arithmetic basically wins here.
+ */
+
+#define fname __strnlen_user
+
+ .text
+ .global fname
+ .type fname, @function
+ .p2align 5 /* why? */
+fname:
+ {
+ mod8 = and(isrc,#7);
+ end = add(isrc,max);
+ start = isrc;
+ }
+ {
+ P0 = cmp.eq(mod8,#0);
+ mod8 = and(end,#7);
+ dcmp = #0;
+ if (P0.new) jump:t dw_loop; /* fire up the oven */
+ }
+
+alignment_loop:
+fail_1: {
+ tmp1 = memb(start++#1);
+ }
+ {
+ P0 = cmp.eq(tmp1,#0);
+ if (P0.new) jump:nt exit_found;
+ P1 = cmp.gtu(end,start);
+ mod8 = and(start,#7);
+ }
+ {
+ if (!P1) jump exit_error; /* hit the end */
+ P0 = cmp.eq(mod8,#0);
+ }
+ {
+ if (!P0) jump alignment_loop;
+ }
+
+
+
+dw_loop:
+fail_2: {
+ dbuf = memd(start);
+ obo = add(start,#1);
+ }
+ {
+ P0 = vcmpb.eq(dbuf,dcmp);
+ }
+ {
+ tmp1 = P0;
+ P0 = cmp.gtu(end,start);
+ }
+ {
+ tmp1 = ct0(tmp1);
+ mod8 = and(end,#7);
+ if (!P0) jump end_check;
+ }
+ {
+ P0 = cmp.eq(tmp1,#32);
+ if (!P0.new) jump:nt exit_found;
+ if (!P0.new) start = add(obo,tmp1);
+ }
+ {
+ start = add(start,#8);
+ jump dw_loop;
+ } /* might be nice to combine these jumps... */
+
+
+end_check:
+ {
+ P0 = cmp.gt(tmp1,mod8);
+ if (P0.new) jump:nt exit_error; /* neverfound! */
+ start = add(obo,tmp1);
+ }
+
+exit_found:
+ {
+ R0 = sub(start,isrc);
+ jumpr R31;
+ }
+
+exit_error:
+ {
+ R0 = add(max,#1);
+ jumpr R31;
+ }
+
+ /* Uh, what does the "fixup" return here? */
+ .falign
+fix_1:
+ {
+ R0 = #0;
+ jumpr R31;
+ }
+
+ .size fname,.-fname
+
+
+.section __ex_table,"a"
+.long fail_1,fix_1
+.long fail_2,fix_1
+.previous
+
diff --git a/arch/hexagon/mm/uaccess.c b/arch/hexagon/mm/uaccess.c
new file mode 100644
index 0000000..e748108
--- /dev/null
+++ b/arch/hexagon/mm/uaccess.c
@@ -0,0 +1,59 @@
+/*
+ * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+/*
+ * Support for user memory access from kernel. This will
+ * probably be inlined for performance at some point, but
+ * for ease of debug, and to a lesser degree for code size,
+ * we implement here as subroutines.
+ */
+#include <linux/types.h>
+#include <asm/uaccess.h>
+#include <asm/pgtable.h>
+
+/*
+ * For clear_user(), exploit previously defined copy_to_user function
+ * and the fact that we've got a handy zero page defined in kernel/head.S
+ *
+ * dczero here would be even faster.
+ */
+__kernel_size_t __clear_user_hexagon(void __user *dest, unsigned long count)
+{
+ long uncleared;
+
+ while (count > PAGE_SIZE) {
+ uncleared = __copy_to_user_hexagon(dest, &empty_zero_page,
+ PAGE_SIZE);
+ if (uncleared)
+ return count - (PAGE_SIZE - uncleared);
+ count -= PAGE_SIZE;
+ dest += PAGE_SIZE;
+ }
+ if (count)
+ count = __copy_to_user_hexagon(dest, &empty_zero_page, count);
+
+ return count;
+}
+
+unsigned long clear_user_hexagon(void __user *dest, unsigned long count)
+{
+ if (!access_ok(VERIFY_WRITE, dest, count))
+ return count;
+ else
+ return __clear_user_hexagon(dest, count);
+}
--
1.7.1
--
Sent by an employee of the Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum.
^ permalink raw reply related
* [PATCH v5 20/36] Hexagon: Provide basic debugging and system trap support.
From: Richard Kuo @ 2011-10-31 23:44 UTC (permalink / raw)
To: torvalds, linux-kernel
Signed-off-by: Richard Kuo <rkuo@codeaurora.org>
Signed-off-by: Linas Vepstas <linas@codeaurora.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
---
arch/hexagon/include/asm/traps.h | 29 +++
arch/hexagon/kernel/traps.c | 454 ++++++++++++++++++++++++++++++++++++++
2 files changed, 483 insertions(+), 0 deletions(-)
create mode 100644 arch/hexagon/include/asm/traps.h
create mode 100644 arch/hexagon/kernel/traps.c
diff --git a/arch/hexagon/include/asm/traps.h b/arch/hexagon/include/asm/traps.h
new file mode 100644
index 0000000..6a407f6
--- /dev/null
+++ b/arch/hexagon/include/asm/traps.h
@@ -0,0 +1,29 @@
+/*
+ * Trap support for Hexagon
+ *
+ * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+#ifndef _ASM_HEXAGON_TRAPS_H
+#define _ASM_HEXAGON_TRAPS_H
+
+#include <asm/registers.h>
+
+extern int die(const char *str, struct pt_regs *regs, long err);
+extern int die_if_kernel(char *str, struct pt_regs *regs, long err);
+
+#endif /* _ASM_HEXAGON_TRAPS_H */
diff --git a/arch/hexagon/kernel/traps.c b/arch/hexagon/kernel/traps.c
new file mode 100644
index 0000000..f08857d
--- /dev/null
+++ b/arch/hexagon/kernel/traps.c
@@ -0,0 +1,454 @@
+/*
+ * Kernel traps/events for Hexagon processor
+ *
+ * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+#include <linux/init.h>
+#include <linux/sched.h>
+#include <linux/module.h>
+#include <linux/kallsyms.h>
+#include <linux/kdebug.h>
+#include <linux/syscalls.h>
+#include <linux/signal.h>
+#include <linux/tracehook.h>
+#include <asm/traps.h>
+#include <asm/vm_fault.h>
+#include <asm/syscall.h>
+#include <asm/registers.h>
+#include <asm/unistd.h>
+#include <asm/sections.h>
+#ifdef CONFIG_KGDB
+# include <linux/kgdb.h>
+#endif
+
+#define TRAP_SYSCALL 1
+#define TRAP_DEBUG 0xdb
+
+void __init trap_init(void)
+{
+}
+
+#ifdef CONFIG_GENERIC_BUG
+/* Maybe should resemble arch/sh/kernel/traps.c ?? */
+int is_valid_bugaddr(unsigned long addr)
+{
+ return 1;
+}
+#endif /* CONFIG_GENERIC_BUG */
+
+static const char *ex_name(int ex)
+{
+ switch (ex) {
+ case HVM_GE_C_XPROT:
+ case HVM_GE_C_XUSER:
+ return "Execute protection fault";
+ case HVM_GE_C_RPROT:
+ case HVM_GE_C_RUSER:
+ return "Read protection fault";
+ case HVM_GE_C_WPROT:
+ case HVM_GE_C_WUSER:
+ return "Write protection fault";
+ case HVM_GE_C_XMAL:
+ return "Misaligned instruction";
+ case HVM_GE_C_RMAL:
+ return "Misaligned data load";
+ case HVM_GE_C_WMAL:
+ return "Misaligned data store";
+ case HVM_GE_C_INVI:
+ case HVM_GE_C_PRIVI:
+ return "Illegal instruction";
+ case HVM_GE_C_BUS:
+ return "Precise bus error";
+ case HVM_GE_C_CACHE:
+ return "Cache error";
+
+ case 0xdb:
+ return "Debugger trap";
+
+ default:
+ return "Unrecognized exception";
+ }
+}
+
+static void do_show_stack(struct task_struct *task, unsigned long *fp,
+ unsigned long ip)
+{
+ int kstack_depth_to_print = 24;
+ unsigned long offset, size;
+ const char *name = NULL;
+ unsigned long *newfp;
+ unsigned long low, high;
+ char tmpstr[128];
+ char *modname;
+ int i;
+
+ if (task == NULL)
+ task = current;
+
+ printk(KERN_INFO "CPU#%d, %s/%d, Call Trace:\n",
+ raw_smp_processor_id(), task->comm,
+ task_pid_nr(task));
+
+ if (fp == NULL) {
+ if (task == current) {
+ asm("%0 = r30" : "=r" (fp));
+ } else {
+ fp = (unsigned long *)
+ ((struct hexagon_switch_stack *)
+ task->thread.switch_sp)->fp;
+ }
+ }
+
+ if ((((unsigned long) fp) & 0x3) || ((unsigned long) fp < 0x1000)) {
+ printk(KERN_INFO "-- Corrupt frame pointer %p\n", fp);
+ return;
+ }
+
+ /* Saved link reg is one word above FP */
+ if (!ip)
+ ip = *(fp+1);
+
+ /* Expect kernel stack to be in-bounds */
+ low = (unsigned long)task_stack_page(task);
+ high = low + THREAD_SIZE - 8;
+ low += sizeof(struct thread_info);
+
+ for (i = 0; i < kstack_depth_to_print; i++) {
+
+ name = kallsyms_lookup(ip, &size, &offset, &modname, tmpstr);
+
+ printk(KERN_INFO "[%p] 0x%lx: %s + 0x%lx", fp, ip, name,
+ offset);
+ if (((unsigned long) fp < low) || (high < (unsigned long) fp))
+ printk(KERN_CONT " (FP out of bounds!)");
+ if (modname)
+ printk(KERN_CONT " [%s] ", modname);
+ printk(KERN_CONT "\n");
+
+ newfp = (unsigned long *) *fp;
+
+ if (((unsigned long) newfp) & 0x3) {
+ printk(KERN_INFO "-- Corrupt frame pointer %p\n",
+ newfp);
+ break;
+ }
+
+ /* Attempt to continue past exception. */
+ if (0 == newfp) {
+ struct pt_regs *regs = (struct pt_regs *) (((void *)fp)
+ + 8);
+
+ if (regs->syscall_nr != -1) {
+ printk(KERN_INFO "-- trap0 -- syscall_nr: %ld",
+ regs->syscall_nr);
+ printk(KERN_CONT " psp: %lx elr: %lx\n",
+ pt_psp(regs), pt_elr(regs));
+ break;
+ } else {
+ /* really want to see more ... */
+ kstack_depth_to_print += 6;
+ printk(KERN_INFO "-- %s (0x%lx) badva: %lx\n",
+ ex_name(pt_cause(regs)), pt_cause(regs),
+ pt_badva(regs));
+ }
+
+ newfp = (unsigned long *) regs->r30;
+ ip = pt_elr(regs);
+ } else {
+ ip = *(newfp + 1);
+ }
+
+ /* If link reg is null, we are done. */
+ if (ip == 0x0)
+ break;
+
+ /* If newfp isn't larger, we're tracing garbage. */
+ if (newfp > fp)
+ fp = newfp;
+ else
+ break;
+ }
+}
+
+void show_stack(struct task_struct *task, unsigned long *fp)
+{
+ /* Saved link reg is one word above FP */
+ do_show_stack(task, fp, 0);
+}
+
+void dump_stack(void)
+{
+ unsigned long *fp;
+ asm("%0 = r30" : "=r" (fp));
+ show_stack(current, fp);
+}
+EXPORT_SYMBOL(dump_stack);
+
+int die(const char *str, struct pt_regs *regs, long err)
+{
+ static struct {
+ spinlock_t lock;
+ int counter;
+ } die = {
+ .lock = __SPIN_LOCK_UNLOCKED(die.lock),
+ .counter = 0
+ };
+
+ console_verbose();
+ oops_enter();
+
+ spin_lock_irq(&die.lock);
+ bust_spinlocks(1);
+ printk(KERN_EMERG "Oops: %s[#%d]:\n", str, ++die.counter);
+
+ if (notify_die(DIE_OOPS, str, regs, err, pt_cause(regs), SIGSEGV) ==
+ NOTIFY_STOP)
+ return 1;
+
+ print_modules();
+ show_regs(regs);
+ do_show_stack(current, ®s->r30, pt_elr(regs));
+
+ bust_spinlocks(0);
+ add_taint(TAINT_DIE);
+
+ spin_unlock_irq(&die.lock);
+
+ if (in_interrupt())
+ panic("Fatal exception in interrupt");
+
+ if (panic_on_oops)
+ panic("Fatal exception");
+
+ oops_exit();
+ do_exit(err);
+ return 0;
+}
+
+int die_if_kernel(char *str, struct pt_regs *regs, long err)
+{
+ if (!user_mode(regs))
+ return die(str, regs, err);
+ else
+ return 0;
+}
+
+/*
+ * It's not clear that misaligned fetches are ever recoverable.
+ */
+static void misaligned_instruction(struct pt_regs *regs)
+{
+ die_if_kernel("Misaligned Instruction", regs, 0);
+ force_sig(SIGBUS, current);
+}
+
+/*
+ * Misaligned loads and stores, on the other hand, can be
+ * emulated, and probably should be, some day. But for now
+ * they will be considered fatal.
+ */
+static void misaligned_data_load(struct pt_regs *regs)
+{
+ die_if_kernel("Misaligned Data Load", regs, 0);
+ force_sig(SIGBUS, current);
+}
+
+static void misaligned_data_store(struct pt_regs *regs)
+{
+ die_if_kernel("Misaligned Data Store", regs, 0);
+ force_sig(SIGBUS, current);
+}
+
+static void illegal_instruction(struct pt_regs *regs)
+{
+ die_if_kernel("Illegal Instruction", regs, 0);
+ force_sig(SIGILL, current);
+}
+
+/*
+ * Precise bus errors may be recoverable with a a retry,
+ * but for now, treat them as irrecoverable.
+ */
+static void precise_bus_error(struct pt_regs *regs)
+{
+ die_if_kernel("Precise Bus Error", regs, 0);
+ force_sig(SIGBUS, current);
+}
+
+/*
+ * If anything is to be done here other than panic,
+ * it will probably be complex and migrate to another
+ * source module. For now, just die.
+ */
+static void cache_error(struct pt_regs *regs)
+{
+ die("Cache Error", regs, 0);
+}
+
+/*
+ * General exception handler
+ */
+void do_genex(struct pt_regs *regs)
+{
+ /*
+ * Decode Cause and Dispatch
+ */
+ switch (pt_cause(regs)) {
+ case HVM_GE_C_XPROT:
+ case HVM_GE_C_XUSER:
+ execute_protection_fault(regs);
+ break;
+ case HVM_GE_C_RPROT:
+ case HVM_GE_C_RUSER:
+ read_protection_fault(regs);
+ break;
+ case HVM_GE_C_WPROT:
+ case HVM_GE_C_WUSER:
+ write_protection_fault(regs);
+ break;
+ case HVM_GE_C_XMAL:
+ misaligned_instruction(regs);
+ break;
+ case HVM_GE_C_RMAL:
+ misaligned_data_load(regs);
+ break;
+ case HVM_GE_C_WMAL:
+ misaligned_data_store(regs);
+ break;
+ case HVM_GE_C_INVI:
+ case HVM_GE_C_PRIVI:
+ illegal_instruction(regs);
+ break;
+ case HVM_GE_C_BUS:
+ precise_bus_error(regs);
+ break;
+ case HVM_GE_C_CACHE:
+ cache_error(regs);
+ break;
+ default:
+ /* Halt and catch fire */
+ panic("Unrecognized exception 0x%lx\n", pt_cause(regs));
+ break;
+ }
+}
+
+/* Indirect system call dispatch */
+long sys_syscall(void)
+{
+ printk(KERN_ERR "sys_syscall invoked!\n");
+ return -ENOSYS;
+}
+
+void do_trap0(struct pt_regs *regs)
+{
+ unsigned long syscallret = 0;
+ syscall_fn syscall;
+
+ switch (pt_cause(regs)) {
+ case TRAP_SYSCALL:
+ /* System call is trap0 #1 */
+
+ /* allow strace to catch syscall args */
+ if (unlikely(test_thread_flag(TIF_SYSCALL_TRACE) &&
+ tracehook_report_syscall_entry(regs)))
+ return; /* return -ENOSYS somewhere? */
+
+ /* Interrupts should be re-enabled for syscall processing */
+ __vmsetie(VM_INT_ENABLE);
+
+ /*
+ * System call number is in r6, arguments in r0..r5.
+ * Fortunately, no Linux syscall has more than 6 arguments,
+ * and Hexagon ABI passes first 6 arguments in registers.
+ * 64-bit arguments are passed in odd/even register pairs.
+ * Fortunately, we have no system calls that take more
+ * than three arguments with more than one 64-bit value.
+ * Should that change, we'd need to redesign to copy
+ * between user and kernel stacks.
+ */
+ regs->syscall_nr = regs->r06;
+
+ /*
+ * GPR R0 carries the first parameter, and is also used
+ * to report the return value. We need a backup of
+ * the user's value in case we need to do a late restart
+ * of the system call.
+ */
+ regs->restart_r0 = regs->r00;
+
+ if ((unsigned long) regs->syscall_nr >= __NR_syscalls) {
+ regs->r00 = -1;
+ } else {
+ syscall = (syscall_fn)
+ (sys_call_table[regs->syscall_nr]);
+ syscallret = syscall(regs->r00, regs->r01,
+ regs->r02, regs->r03,
+ regs->r04, regs->r05);
+ }
+
+ /*
+ * If it was a sigreturn system call, don't overwrite
+ * r0 value in stack frame with return value.
+ *
+ * __NR_sigreturn doesn't seem to exist in new unistd.h
+ */
+
+ if (regs->syscall_nr != __NR_rt_sigreturn)
+ regs->r00 = syscallret;
+
+ /* allow strace to get the syscall return state */
+ if (unlikely(test_thread_flag(TIF_SYSCALL_TRACE)))
+ tracehook_report_syscall_exit(regs, 0);
+
+ break;
+ case TRAP_DEBUG:
+ /* Trap0 0xdb is debug breakpoint */
+ if (user_mode(regs)) {
+ struct siginfo info;
+
+ info.si_signo = SIGTRAP;
+ info.si_errno = 0;
+ /*
+ * Some architecures add some per-thread state
+ * to distinguish between breakpoint traps and
+ * trace traps. We may want to do that, and
+ * set the si_code value appropriately, or we
+ * may want to use a different trap0 flavor.
+ */
+ info.si_code = TRAP_BRKPT;
+ info.si_addr = (void __user *) pt_elr(regs);
+ send_sig_info(SIGTRAP, &info, current);
+ } else {
+#ifdef CONFIG_KGDB
+ kgdb_handle_exception(pt_cause(regs), SIGTRAP,
+ TRAP_BRKPT, regs);
+#endif
+ }
+ break;
+ }
+ /* Ignore other trap0 codes for now, especially 0 (Angel calls) */
+}
+
+/*
+ * Machine check exception handler
+ */
+void do_machcheck(struct pt_regs *regs)
+{
+ /* Halt and catch fire */
+ __vmstop();
+}
--
1.7.1
--
Sent by an employee of the Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum.
^ permalink raw reply related
* [PATCH v5 16/36] Hexagon: Add startup code
From: Richard Kuo @ 2011-10-31 23:42 UTC (permalink / raw)
To: torvalds, linux-kernel
Signed-off-by: Richard Kuo <rkuo@codeaurora.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
---
arch/hexagon/kernel/head.S | 162 +++++++++++++++++++++++++++++++++++++++++++
arch/hexagon/kernel/reset.c | 38 ++++++++++
arch/hexagon/kernel/setup.c | 145 ++++++++++++++++++++++++++++++++++++++
3 files changed, 345 insertions(+), 0 deletions(-)
create mode 100644 arch/hexagon/kernel/head.S
create mode 100644 arch/hexagon/kernel/reset.c
create mode 100644 arch/hexagon/kernel/setup.c
diff --git a/arch/hexagon/kernel/head.S b/arch/hexagon/kernel/head.S
new file mode 100644
index 0000000..6964da5
--- /dev/null
+++ b/arch/hexagon/kernel/head.S
@@ -0,0 +1,162 @@
+/*
+ * Early kernel startup code for Hexagon
+ *
+ * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved.
+ *
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+#include <linux/linkage.h>
+#include <linux/init.h>
+#include <asm/asm-offsets.h>
+#include <asm/mem-layout.h>
+#include <asm/vm_mmu.h>
+#include <asm/page.h>
+
+ __INIT
+ENTRY(stext)
+ /*
+ * VMM will already have set up true vector page, MMU, etc.
+ * To set up initial kernel identity map, we have to pass
+ * the VMM a pointer to some canonical page tables. In
+ * this implementation, we're assuming that we've got
+ * them precompiled. Generate value in R24, as we'll need
+ * it again shortly.
+ */
+ r24.L = #LO(swapper_pg_dir)
+ r24.H = #HI(swapper_pg_dir)
+
+ /*
+ * Symbol is kernel segment address, but we need
+ * the logical/physical address.
+ */
+ r24 = asl(r24, #2)
+ r24 = lsr(r24, #2)
+
+ r0 = r24
+
+ /*
+ * Initialize a 16MB PTE to make the virtual and physical
+ * addresses where the kernel was loaded be identical.
+ */
+#define PTE_BITS ( __HVM_PTE_R | __HVM_PTE_W | __HVM_PTE_X \
+ | __HEXAGON_C_WB_L2 << 6 \
+ | __HVM_PDE_S_4MB)
+
+ r1 = pc
+ r2.H = #0xffc0
+ r2.L = #0x0000
+ r1 = and(r1,r2) /* round PC to 4MB boundary */
+ r2 = lsr(r1, #22) /* 4MB page number */
+ r2 = asl(r2, #2) /* times sizeof(PTE) (4bytes) */
+ r0 = add(r0,r2) /* r0 = address of correct PTE */
+ r2 = #PTE_BITS
+ r1 = add(r1,r2) /* r1 = 4MB PTE for the first entry */
+ r2.h = #0x0040
+ r2.l = #0x0000 /* 4MB */
+ memw(r0 ++ #4) = r1
+ r1 = add(r1, r2)
+ memw(r0 ++ #4) = r1
+
+ r0 = r24
+
+ /*
+ * The subroutine wrapper around the virtual instruction touches
+ * no memory, so we should be able to use it even here.
+ */
+ call __vmnewmap;
+
+ /* Jump into virtual address range. */
+
+ r31.h = #hi(__head_s_vaddr_target)
+ r31.l = #lo(__head_s_vaddr_target)
+ jumpr r31
+
+ /* Insert trippy space effects. */
+
+__head_s_vaddr_target:
+ /*
+ * Tear down VA=PA translation now that we are running
+ * in the desgnated kernel segments.
+ */
+ r0 = #__HVM_PDE_S_INVALID
+ r1 = r24
+ loop0(1f,#0x100)
+1:
+ {
+ memw(R1 ++ #4) = R0
+ }:endloop0
+
+ r0 = r24
+ call __vmnewmap
+
+ /* Go ahead and install the trap0 return so angel calls work */
+ r0.h = #hi(_K_provisional_vec)
+ r0.l = #lo(_K_provisional_vec)
+ call __vmsetvec
+
+ /*
+ * OK, at this point we should start to be much more careful,
+ * we're going to enter C code and start touching memory
+ * in all sorts of places.
+ * This means:
+ * SGP needs to be OK
+ * Need to lock shared resources
+ * A bunch of other things that will cause
+ * all kinds of painful bugs
+ */
+
+ /*
+ * Stack pointer should be pointed at the init task's
+ * thread stack, which should have been declared in arch/init_task.c.
+ * So uhhhhh...
+ * It's accessible via the init_thread_union, which is a union
+ * of a thread_info struct and a stack; of course, the top
+ * of the stack is not for you. The end of the stack
+ * is simply init_thread_union + THREAD_SIZE.
+ */
+
+ {r29.H = #HI(init_thread_union); r0.H = #HI(_THREAD_SIZE); }
+ {r29.L = #LO(init_thread_union); r0.L = #LO(_THREAD_SIZE); }
+
+ /* initialize the register used to point to current_thread_info */
+ /* Fixme: THREADINFO_REG can't be R2 because of that memset thing. */
+ {r29 = add(r29,r0); THREADINFO_REG = r29; }
+
+ /* Hack: zero bss; */
+ { r0.L = #LO(__bss_start); r1 = #0; r2.l = #LO(__bss_stop); }
+ { r0.H = #HI(__bss_start); r2.h = #HI(__bss_stop); }
+
+ r2 = sub(r2,r0);
+ call memset;
+
+ /* Time to make the doughnuts. */
+ call start_kernel
+
+ /*
+ * Should not reach here.
+ */
+1:
+ jump 1b
+
+.p2align PAGE_SHIFT
+ENTRY(external_cmdline_buffer)
+ .fill _PAGE_SIZE,1,0
+
+.data
+.p2align PAGE_SHIFT
+ENTRY(empty_zero_page)
+ .fill _PAGE_SIZE,1,0
diff --git a/arch/hexagon/kernel/reset.c b/arch/hexagon/kernel/reset.c
new file mode 100644
index 0000000..4d72fc5
--- /dev/null
+++ b/arch/hexagon/kernel/reset.c
@@ -0,0 +1,38 @@
+/*
+ * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+#include <linux/smp.h>
+#include <asm/hexagon_vm.h>
+
+void machine_power_off(void)
+{
+ smp_send_stop();
+ __vmstop();
+}
+
+void machine_halt(void)
+{
+}
+
+void machine_restart(char *cmd)
+{
+}
+
+void pm_power_off(void)
+{
+}
diff --git a/arch/hexagon/kernel/setup.c b/arch/hexagon/kernel/setup.c
new file mode 100644
index 0000000..1202f78
--- /dev/null
+++ b/arch/hexagon/kernel/setup.c
@@ -0,0 +1,145 @@
+/*
+ * Arch related setup for Hexagon
+ *
+ * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+#include <linux/init.h>
+#include <linux/bootmem.h>
+#include <linux/mmzone.h>
+#include <linux/mm.h>
+#include <linux/seq_file.h>
+#include <linux/console.h>
+#include <linux/of_fdt.h>
+#include <asm/io.h>
+#include <asm/sections.h>
+#include <asm/setup.h>
+#include <asm/processor.h>
+#include <asm/hexagon_vm.h>
+#include <asm/vm_mmu.h>
+#include <asm/time.h>
+#ifdef CONFIG_OF
+#include <asm/prom.h>
+#endif
+
+char cmd_line[COMMAND_LINE_SIZE];
+static char default_command_line[COMMAND_LINE_SIZE] __initdata = CONFIG_CMDLINE;
+
+int on_simulator;
+
+void __cpuinit calibrate_delay(void)
+{
+ loops_per_jiffy = thread_freq_mhz * 1000000 / HZ;
+}
+
+/*
+ * setup_arch - high level architectural setup routine
+ * @cmdline_p: pointer to pointer to command-line arguments
+ */
+
+void __init setup_arch(char **cmdline_p)
+{
+ char *p = &external_cmdline_buffer;
+
+ /*
+ * These will eventually be pulled in via either some hypervisor
+ * or devicetree description. Hardwiring for now.
+ */
+ pcycle_freq_mhz = 600;
+ thread_freq_mhz = 100;
+ sleep_clk_freq = 32000;
+
+ /*
+ * Set up event bindings to handle exceptions and interrupts.
+ */
+ __vmsetvec(_K_VM_event_vector);
+
+ /*
+ * Simulator has a few differences from the hardware.
+ * For now, check uninitialized-but-mapped memory
+ * prior to invoking setup_arch_memory().
+ */
+ if (*(int *)((unsigned long)_end + 8) == 0x1f1f1f1f)
+ on_simulator = 1;
+ else
+ on_simulator = 0;
+
+ if (p[0] != '\0')
+ strlcpy(boot_command_line, p, COMMAND_LINE_SIZE);
+ else
+ strlcpy(boot_command_line, default_command_line,
+ COMMAND_LINE_SIZE);
+
+ /*
+ * boot_command_line and the value set up by setup_arch
+ * are both picked up by the init code. If no reason to
+ * make them different, pass the same pointer back.
+ */
+ strlcpy(cmd_line, boot_command_line, COMMAND_LINE_SIZE);
+ *cmdline_p = cmd_line;
+
+ parse_early_param();
+
+ setup_arch_memory();
+
+#ifdef CONFIG_SMP
+ smp_start_cpus();
+#endif
+}
+
+/*
+ * Functions for dumping CPU info via /proc
+ * Probably should move to kernel/proc.c or something.
+ */
+static void *c_start(struct seq_file *m, loff_t *pos)
+{
+ return *pos < nr_cpu_ids ? (void *)((unsigned long) *pos + 1) : NULL;
+}
+
+static void *c_next(struct seq_file *m, void *v, loff_t *pos)
+{
+ ++*pos;
+ return c_start(m, pos);
+}
+
+static void c_stop(struct seq_file *m, void *v)
+{
+}
+
+/*
+ * Eventually this will dump information about
+ * CPU properties like ISA level, TLB size, etc.
+ */
+static int show_cpuinfo(struct seq_file *m, void *v)
+{
+ int cpu = (unsigned long) v - 1;
+
+ seq_printf(m, "processor\t: %d\n", cpu);
+ seq_printf(m, "model name\t: Hexagon Virtual Machine\n");
+ seq_printf(m, "BogoMips\t: %lu.%02lu\n",
+ (loops_per_jiffy * HZ) / 500000,
+ ((loops_per_jiffy * HZ) / 5000) % 100);
+ seq_printf(m, "\n");
+ return 0;
+}
+
+const struct seq_operations cpuinfo_op = {
+ .start = &c_start,
+ .next = &c_next,
+ .stop = &c_stop,
+ .show = &show_cpuinfo,
+};
--
1.7.1
--
Sent by an employee of the Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum.
^ permalink raw reply related
* [PATCH v5 17/36] Hexagon: Add interrupts
From: Richard Kuo @ 2011-10-31 23:42 UTC (permalink / raw)
To: torvalds, linux-kernel
Signed-off-by: Richard Kuo <rkuo@codeaurora.org>
Signed-off-by: Linas Vepstas <linas@codeaurora.org>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Arnd Bergmann <arnd@arndb.de>
---
arch/hexagon/include/asm/irq.h | 36 ++++++++++++++++
arch/hexagon/kernel/irq_cpu.c | 90 ++++++++++++++++++++++++++++++++++++++++
2 files changed, 126 insertions(+), 0 deletions(-)
create mode 100644 arch/hexagon/include/asm/irq.h
create mode 100644 arch/hexagon/kernel/irq_cpu.c
diff --git a/arch/hexagon/include/asm/irq.h b/arch/hexagon/include/asm/irq.h
new file mode 100644
index 0000000..ded8c15
--- /dev/null
+++ b/arch/hexagon/include/asm/irq.h
@@ -0,0 +1,36 @@
+/*
+ * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+#ifndef _ASM_IRQ_H_
+#define _ASM_IRQ_H_
+
+/* Number of first-level interrupts associated with the CPU core. */
+#define HEXAGON_CPUINTS 32
+
+/*
+ * Must define NR_IRQS before including <asm-generic/irq.h>
+ * 64 == the two SIRC's, 176 == the two gpio's
+ *
+ * IRQ configuration is still in flux; defining this to a comfortably
+ * large number.
+ */
+#define NR_IRQS 512
+
+#include <asm-generic/irq.h>
+
+#endif
diff --git a/arch/hexagon/kernel/irq_cpu.c b/arch/hexagon/kernel/irq_cpu.c
new file mode 100644
index 0000000..d4416a1
--- /dev/null
+++ b/arch/hexagon/kernel/irq_cpu.c
@@ -0,0 +1,90 @@
+/*
+ * First-level interrupt controller model for Hexagon.
+ *
+ * Copyright (c) 2010-2011 Code Aurora Forum. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+#include <linux/interrupt.h>
+#include <asm/irq.h>
+#include <asm/hexagon_vm.h>
+
+static void mask_irq(struct irq_data *data)
+{
+ __vmintop_locdis((long) data->irq);
+}
+
+static void mask_irq_num(unsigned int irq)
+{
+ __vmintop_locdis((long) irq);
+}
+
+static void unmask_irq(struct irq_data *data)
+{
+ __vmintop_locen((long) data->irq);
+}
+
+/* This is actually all we need for handle_fasteoi_irq */
+static void eoi_irq(struct irq_data *data)
+{
+ __vmintop_globen((long) data->irq);
+}
+
+/* Power mamangement wake call. We don't need this, however,
+ * if this is absent, then an -ENXIO error is returned to the
+ * msm_serial driver, and it fails to correctly initialize.
+ * This is a bug in the msm_serial driver, but, for now, we
+ * work around it here, by providing this bogus handler.
+ * XXX FIXME!!! remove this when msm_serial is fixed.
+ */
+static int set_wake(struct irq_data *data, unsigned int on)
+{
+ return 0;
+}
+
+static struct irq_chip hexagon_irq_chip = {
+ .name = "HEXAGON",
+ .irq_mask = mask_irq,
+ .irq_unmask = unmask_irq,
+ .irq_set_wake = set_wake,
+ .irq_eoi = eoi_irq
+};
+
+/**
+ * The hexagon core comes with a first-level interrupt controller
+ * with 32 total possible interrupts. When the core is embedded
+ * into different systems/platforms, it is typically wrapped by
+ * macro cells that provide one or more second-level interrupt
+ * controllers that are cascaded into one or more of the first-level
+ * interrupts handled here. The precise wiring of these other
+ * irqs varies from platform to platform, and are set up & configured
+ * in the platform-specific files.
+ *
+ * The first-level interrupt controller is wrapped by the VM, which
+ * virtualizes the interrupt controller for us. It provides a very
+ * simple, fast & efficient API, and so the fasteoi handler is
+ * appropriate for this case.
+ */
+void __init init_IRQ(void)
+{
+ int irq;
+
+ for (irq = 0; irq < HEXAGON_CPUINTS; irq++) {
+ mask_irq_num(irq);
+ irq_set_chip_and_handler(irq, &hexagon_irq_chip,
+ handle_fasteoi_irq);
+ }
+}
--
1.7.1
--
Sent by an employee of the Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum.
^ permalink raw reply related
* [PATCH v5 15/36] Hexagon: Add init_task and process functions
From: Richard Kuo @ 2011-10-31 23:41 UTC (permalink / raw)
To: torvalds, linux-kernel
Signed-off-by: Richard Kuo <rkuo@codeaurora.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
---
arch/hexagon/kernel/init_task.c | 54 ++++++++
arch/hexagon/kernel/process.c | 279 +++++++++++++++++++++++++++++++++++++++
2 files changed, 333 insertions(+), 0 deletions(-)
create mode 100644 arch/hexagon/kernel/init_task.c
create mode 100644 arch/hexagon/kernel/process.c
diff --git a/arch/hexagon/kernel/init_task.c b/arch/hexagon/kernel/init_task.c
new file mode 100644
index 0000000..73283d3
--- /dev/null
+++ b/arch/hexagon/kernel/init_task.c
@@ -0,0 +1,54 @@
+/*
+ * Init task definition
+ *
+ * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+#include <linux/mm.h>
+#include <linux/module.h>
+#include <linux/sched.h>
+#include <linux/init_task.h>
+#include <linux/fs.h>
+#include <linux/mqueue.h>
+#include <asm/thread_info.h>
+#include <asm/uaccess.h>
+#include <asm/pgtable.h>
+
+static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
+static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
+
+/*
+ * Initial thread structure.
+ *
+ * We need to make sure that this is 8192-byte aligned due to the
+ * way process stacks are handled. This is done by making sure
+ * the linker maps this in the .text segment right after head.S,
+ * and making head.S ensure the proper alignment.
+ */
+union thread_union init_thread_union
+ __attribute__((__section__(".data.init_task"),
+ __aligned__(THREAD_SIZE))) = {
+ INIT_THREAD_INFO(init_task)
+ };
+
+/*
+ * Initial task structure.
+ *
+ * All other task structs will be allocated on slabs in fork.c
+ */
+struct task_struct init_task = INIT_TASK(init_task);
+EXPORT_SYMBOL(init_task);
diff --git a/arch/hexagon/kernel/process.c b/arch/hexagon/kernel/process.c
new file mode 100644
index 0000000..18c4f0b
--- /dev/null
+++ b/arch/hexagon/kernel/process.c
@@ -0,0 +1,279 @@
+/*
+ * Process creation support for Hexagon
+ *
+ * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+#include <linux/sched.h>
+#include <linux/types.h>
+#include <linux/module.h>
+#include <linux/tick.h>
+#include <linux/uaccess.h>
+#include <linux/slab.h>
+
+/*
+ * Kernel thread creation. The desired kernel function is "wrapped"
+ * in the kernel_thread_helper function, which does cleanup
+ * afterwards.
+ */
+static void __noreturn kernel_thread_helper(void *arg, int (*fn)(void *))
+{
+ do_exit(fn(arg));
+}
+
+int kernel_thread(int (*fn)(void *), void *arg, unsigned long flags)
+{
+ struct pt_regs regs;
+
+ memset(®s, 0, sizeof(regs));
+ /*
+ * Yes, we're exploting illicit knowledge of the ABI here.
+ */
+ regs.r00 = (unsigned long) arg;
+ regs.r01 = (unsigned long) fn;
+ pt_set_elr(®s, (unsigned long)kernel_thread_helper);
+ pt_set_kmode(®s);
+
+ return do_fork(flags|CLONE_VM|CLONE_UNTRACED, 0, ®s, 0, NULL, NULL);
+}
+EXPORT_SYMBOL(kernel_thread);
+
+/*
+ * Program thread launch. Often defined as a macro in processor.h,
+ * but we're shooting for a small footprint and it's not an inner-loop
+ * performance-critical operation.
+ *
+ * The Hexagon ABI specifies that R28 is zero'ed before program launch,
+ * so that gets automatically done here. If we ever stop doing that here,
+ * we'll probably want to define the ELF_PLAT_INIT macro.
+ */
+void start_thread(struct pt_regs *regs, unsigned long pc, unsigned long sp)
+{
+ /* Set to run with user-mode data segmentation */
+ set_fs(USER_DS);
+ /* We want to zero all data-containing registers. Is this overkill? */
+ memset(regs, 0, sizeof(*regs));
+ /* We might want to also zero all Processor registers here */
+ pt_set_usermode(regs);
+ pt_set_elr(regs, pc);
+ pt_set_rte_sp(regs, sp);
+}
+
+/*
+ * Spin, or better still, do a hardware or VM wait instruction
+ * If hardware or VM offer wait termination even though interrupts
+ * are disabled.
+ */
+static void default_idle(void)
+{
+ __vmwait();
+}
+
+void (*idle_sleep)(void) = default_idle;
+
+void cpu_idle(void)
+{
+ while (1) {
+ tick_nohz_stop_sched_tick(1);
+ local_irq_disable();
+ while (!need_resched()) {
+ idle_sleep();
+ /* interrupts wake us up, but aren't serviced */
+ local_irq_enable(); /* service interrupt */
+ local_irq_disable();
+ }
+ local_irq_enable();
+ tick_nohz_restart_sched_tick();
+ schedule();
+ }
+}
+
+/*
+ * Return saved PC of a blocked thread
+ */
+unsigned long thread_saved_pc(struct task_struct *tsk)
+{
+ return 0;
+}
+
+/*
+ * Copy architecture-specific thread state
+ */
+int copy_thread(unsigned long clone_flags, unsigned long usp,
+ unsigned long unused, struct task_struct *p,
+ struct pt_regs *regs)
+{
+ struct thread_info *ti = task_thread_info(p);
+ struct hexagon_switch_stack *ss;
+ struct pt_regs *childregs;
+ asmlinkage void ret_from_fork(void);
+
+ childregs = (struct pt_regs *) (((unsigned long) ti + THREAD_SIZE) -
+ sizeof(*childregs));
+
+ memcpy(childregs, regs, sizeof(*childregs));
+ ti->regs = childregs;
+
+ /*
+ * Establish kernel stack pointer and initial PC for new thread
+ */
+ ss = (struct hexagon_switch_stack *) ((unsigned long) childregs -
+ sizeof(*ss));
+ ss->lr = (unsigned long)ret_from_fork;
+ p->thread.switch_sp = ss;
+
+ /* If User mode thread, set pt_reg stack pointer as per parameter */
+ if (user_mode(childregs)) {
+ pt_set_rte_sp(childregs, usp);
+
+ /* Child sees zero return value */
+ childregs->r00 = 0;
+
+ /*
+ * The clone syscall has the C signature:
+ * int [r0] clone(int flags [r0],
+ * void *child_frame [r1],
+ * void *parent_tid [r2],
+ * void *child_tid [r3],
+ * void *thread_control_block [r4]);
+ * ugp is used to provide TLS support.
+ */
+ if (clone_flags & CLONE_SETTLS)
+ childregs->ugp = childregs->r04;
+
+ /*
+ * Parent sees new pid -- not necessary, not even possible at
+ * this point in the fork process
+ * Might also want to set things like ti->addr_limit
+ */
+ } else {
+ /*
+ * If kernel thread, resume stack is kernel stack base.
+ * Note that this is pointer arithmetic on pt_regs *
+ */
+ pt_set_rte_sp(childregs, (unsigned long)(childregs + 1));
+ /*
+ * We need the current thread_info fast path pointer
+ * set up in pt_regs. The register to be used is
+ * parametric for assembler code, but the mechanism
+ * doesn't drop neatly into C. Needs to be fixed.
+ */
+ childregs->THREADINFO_REG = (unsigned long) ti;
+ }
+
+ /*
+ * thread_info pointer is pulled out of task_struct "stack"
+ * field on switch_to.
+ */
+ p->stack = (void *)ti;
+
+ return 0;
+}
+
+/*
+ * Release any architecture-specific resources locked by thread
+ */
+void release_thread(struct task_struct *dead_task)
+{
+}
+
+/*
+ * Free any architecture-specific thread data structures, etc.
+ */
+void exit_thread(void)
+{
+}
+
+/*
+ * Some archs flush debug and FPU info here
+ */
+void flush_thread(void)
+{
+}
+
+/*
+ * The "wait channel" terminology is archaic, but what we want
+ * is an identification of the point at which the scheduler
+ * was invoked by a blocked thread.
+ */
+unsigned long get_wchan(struct task_struct *p)
+{
+ unsigned long fp, pc;
+ unsigned long stack_page;
+ int count = 0;
+ if (!p || p == current || p->state == TASK_RUNNING)
+ return 0;
+
+ stack_page = (unsigned long)task_stack_page(p);
+ fp = ((struct hexagon_switch_stack *)p->thread.switch_sp)->fp;
+ do {
+ if (fp < (stack_page + sizeof(struct thread_info)) ||
+ fp >= (THREAD_SIZE - 8 + stack_page))
+ return 0;
+ pc = ((unsigned long *)fp)[1];
+ if (!in_sched_functions(pc))
+ return pc;
+ fp = *(unsigned long *) fp;
+ } while (count++ < 16);
+
+ return 0;
+}
+
+/*
+ * Borrowed from PowerPC -- basically allow smaller kernel stacks if we
+ * go crazy with the page sizes.
+ */
+#if THREAD_SHIFT < PAGE_SHIFT
+
+static struct kmem_cache *thread_info_cache;
+
+struct thread_info *alloc_thread_info_node(struct task_struct *tsk, int node)
+{
+ struct thread_info *ti;
+
+ ti = kmem_cache_alloc_node(thread_info_cache, GFP_KERNEL, node);
+ if (unlikely(ti == NULL))
+ return NULL;
+#ifdef CONFIG_DEBUG_STACK_USAGE
+ memset(ti, 0, THREAD_SIZE);
+#endif
+ return ti;
+}
+
+void free_thread_info(struct thread_info *ti)
+{
+ kmem_cache_free(thread_info_cache, ti);
+}
+
+/* Weak symbol; called by init/main.c */
+
+void thread_info_cache_init(void)
+{
+ thread_info_cache = kmem_cache_create("thread_info", THREAD_SIZE,
+ THREAD_SIZE, 0, NULL);
+ BUG_ON(thread_info_cache == NULL);
+}
+
+#endif /* THREAD_SHIFT < PAGE_SHIFT */
+
+/*
+ * Required placeholder.
+ */
+int dump_fpu(struct pt_regs *regs, elf_fpregset_t *fpu)
+{
+ return 0;
+}
--
1.7.1
--
Sent by an employee of the Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum.
^ permalink raw reply related
* [PATCH v5 13/36] Hexagon: Support dynamic module loading.
From: Richard Kuo @ 2011-10-31 23:40 UTC (permalink / raw)
To: torvalds, linux-kernel
Modules should be compiled as ordinary .o's; shared objects are not
supported.
Signed-off-by: Linas Vepstas <linas@codeaurora.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
---
arch/hexagon/kernel/module.c | 162 ++++++++++++++++++++++++++++++++++++++++++
1 files changed, 162 insertions(+), 0 deletions(-)
create mode 100644 arch/hexagon/kernel/module.c
diff --git a/arch/hexagon/kernel/module.c b/arch/hexagon/kernel/module.c
new file mode 100644
index 0000000..61a76ba
--- /dev/null
+++ b/arch/hexagon/kernel/module.c
@@ -0,0 +1,162 @@
+/*
+ * Kernel module loader for Hexagon
+ *
+ * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+#include <asm/module.h>
+#include <linux/elf.h>
+#include <linux/module.h>
+#include <linux/moduleloader.h>
+#include <linux/vmalloc.h>
+
+#if 0
+#define DEBUGP printk
+#else
+#define DEBUGP(fmt , ...)
+#endif
+
+/*
+ * module_frob_arch_sections - tweak got/plt sections.
+ * @hdr - pointer to elf header
+ * @sechdrs - pointer to elf load section headers
+ * @secstrings - symbol names
+ * @mod - pointer to module
+ */
+int module_frob_arch_sections(Elf_Ehdr *hdr, Elf_Shdr *sechdrs,
+ char *secstrings,
+ struct module *mod)
+{
+ unsigned int i;
+ int found = 0;
+
+ /* Look for .plt and/or .got.plt and/or .init.plt sections */
+ for (i = 0; i < hdr->e_shnum; i++) {
+ DEBUGP("Section %d is %s\n", i,
+ secstrings + sechdrs[i].sh_name);
+ if (strcmp(secstrings + sechdrs[i].sh_name, ".plt") == 0)
+ found = i+1;
+ if (strcmp(secstrings + sechdrs[i].sh_name, ".got.plt") == 0)
+ found = i+1;
+ if (strcmp(secstrings + sechdrs[i].sh_name, ".rela.plt") == 0)
+ found = i+1;
+ }
+
+ /* At this time, we don't support modules comiled with -shared */
+ if (found) {
+ printk(KERN_WARNING
+ "Module '%s' contains unexpected .plt/.got sections.\n",
+ mod->name);
+ /* return -ENOEXEC; */
+ }
+
+ return 0;
+}
+
+/*
+ * apply_relocate_add - perform rela relocations.
+ * @sechdrs - pointer to section headers
+ * @strtab - some sort of start address?
+ * @symindex - symbol index offset or something?
+ * @relsec - address to relocate to?
+ * @module - pointer to module
+ *
+ * Perform rela relocations.
+ */
+int apply_relocate_add(Elf_Shdr *sechdrs, const char *strtab,
+ unsigned int symindex, unsigned int relsec,
+ struct module *module)
+{
+ unsigned int i;
+ Elf32_Sym *sym;
+ uint32_t *location;
+ uint32_t value;
+ unsigned int nrelocs = sechdrs[relsec].sh_size / sizeof(Elf32_Rela);
+ Elf32_Rela *rela = (void *)sechdrs[relsec].sh_addr;
+ Elf32_Word sym_info = sechdrs[relsec].sh_info;
+ Elf32_Sym *sym_base = (Elf32_Sym *) sechdrs[symindex].sh_addr;
+ void *loc_base = (void *) sechdrs[sym_info].sh_addr;
+
+ DEBUGP("Applying relocations in section %u to section %u base=%p\n",
+ relsec, sym_info, loc_base);
+
+ for (i = 0; i < nrelocs; i++) {
+
+ /* Symbol to relocate */
+ sym = sym_base + ELF32_R_SYM(rela[i].r_info);
+
+ /* Where to make the change */
+ location = loc_base + rela[i].r_offset;
+
+ /* `Everything is relative'. */
+ value = sym->st_value + rela[i].r_addend;
+
+ DEBUGP("%d: value=%08x loc=%p reloc=%d symbol=%s\n",
+ i, value, location, ELF32_R_TYPE(rela[i].r_info),
+ sym->st_name ?
+ &strtab[sym->st_name] : "(anonymous)");
+
+ switch (ELF32_R_TYPE(rela[i].r_info)) {
+ case R_HEXAGON_B22_PCREL: {
+ int dist = (int)(value - (uint32_t)location);
+ if ((dist < -0x00800000) ||
+ (dist >= 0x00800000)) {
+ printk(KERN_ERR
+ "%s: %s: %08x=%08x-%08x %s\n",
+ module->name,
+ "R_HEXAGON_B22_PCREL reloc out of range",
+ dist, value, (uint32_t)location,
+ sym->st_name ?
+ &strtab[sym->st_name] : "(anonymous)");
+ return -ENOEXEC;
+ }
+ DEBUGP("B22_PCREL contents: %08X.\n", *location);
+ *location &= ~0x01ff3fff;
+ *location |= 0x00003fff & dist;
+ *location |= 0x01ff0000 & (dist<<2);
+ DEBUGP("Contents after reloc: %08x\n", *location);
+ break;
+ }
+ case R_HEXAGON_HI16:
+ value = (value>>16) & 0xffff;
+ /* fallthrough */
+ case R_HEXAGON_LO16:
+ *location &= ~0x00c03fff;
+ *location |= value & 0x3fff;
+ *location |= (value & 0xc000) << 8;
+ break;
+ case R_HEXAGON_32:
+ *location = value;
+ break;
+ case R_HEXAGON_32_PCREL:
+ *location = value - (uint32_t)location;
+ break;
+ case R_HEXAGON_PLT_B22_PCREL:
+ case R_HEXAGON_GOTOFF_LO16:
+ case R_HEXAGON_GOTOFF_HI16:
+ printk(KERN_ERR "%s: GOT/PLT relocations unsupported\n",
+ module->name);
+ return -ENOEXEC;
+ default:
+ printk(KERN_ERR "%s: unknown relocation: %u\n",
+ module->name,
+ ELF32_R_TYPE(rela[i].r_info));
+ return -ENOEXEC;
+ }
+ }
+ return 0;
+}
--
1.7.1
--
Sent by an employee of the Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum.
^ permalink raw reply related
* [PATCH v5 10/36] Hexagon: Add memcpy and memset accelerated functions
From: Richard Kuo @ 2011-10-31 23:38 UTC (permalink / raw)
To: torvalds, linux-kernel
Signed-off-by: Richard Kuo <rkuo@codeaurora.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
---
arch/hexagon/lib/memcpy.S | 543 +++++++++++++++++++++++++++++++++++++++++++++
arch/hexagon/lib/memset.S | 316 ++++++++++++++++++++++++++
2 files changed, 859 insertions(+), 0 deletions(-)
create mode 100644 arch/hexagon/lib/memcpy.S
create mode 100644 arch/hexagon/lib/memset.S
diff --git a/arch/hexagon/lib/memcpy.S b/arch/hexagon/lib/memcpy.S
new file mode 100644
index 0000000..2101c33
--- /dev/null
+++ b/arch/hexagon/lib/memcpy.S
@@ -0,0 +1,543 @@
+/*
+ * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved.
+ *
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+/*
+ * Description
+ *
+ * library function for memcpy where length bytes are copied from
+ * ptr_in to ptr_out. ptr_out is returned unchanged.
+ * Allows any combination of alignment on input and output pointers
+ * and length from 0 to 2^32-1
+ *
+ * Restrictions
+ * The arrays should not overlap, the program will produce undefined output
+ * if they do.
+ * For blocks less than 16 bytes a byte by byte copy is performed. For
+ * 8byte alignments, and length multiples, a dword copy is performed up to
+ * 96bytes
+ * History
+ *
+ * DJH 5/15/09 Initial version 1.0
+ * DJH 6/ 1/09 Version 1.1 modified ABI to inlcude R16-R19
+ * DJH 7/12/09 Version 1.2 optimized codesize down to 760 was 840
+ * DJH 10/14/09 Version 1.3 added special loop for aligned case, was
+ * overreading bloated codesize back up to 892
+ * DJH 4/20/10 Version 1.4 fixed Ldword_loop_epilog loop to prevent loads
+ * occuring if only 1 left outstanding, fixes bug
+ * # 3888, corrected for all alignments. Peeled off
+ * 1 32byte chunk from kernel loop and extended 8byte
+ * loop at end to solve all combinations and prevent
+ * over read. Fixed Ldword_loop_prolog to prevent
+ * overread for blocks less than 48bytes. Reduced
+ * codesize to 752 bytes
+ * DJH 4/21/10 version 1.5 1.4 fix broke code for input block ends not
+ * aligned to dword boundaries,underwriting by 1
+ * byte, added detection for this and fixed. A
+ * little bloat.
+ * DJH 4/23/10 version 1.6 corrected stack error, R20 was not being restored
+ * always, fixed the error of R20 being modified
+ * before it was being saved
+ * Natural c model
+ * ===============
+ * void * memcpy(char * ptr_out, char * ptr_in, int length) {
+ * int i;
+ * if(length) for(i=0; i < length; i++) { ptr_out[i] = ptr_in[i]; }
+ * return(ptr_out);
+ * }
+ *
+ * Optimized memcpy function
+ * =========================
+ * void * memcpy(char * ptr_out, char * ptr_in, int len) {
+ * int i, prolog, kernel, epilog, mask;
+ * u8 offset;
+ * s64 data0, dataF8, data70;
+ *
+ * s64 * ptr8_in;
+ * s64 * ptr8_out;
+ * s32 * ptr4;
+ * s16 * ptr2;
+ *
+ * offset = ((int) ptr_in) & 7;
+ * ptr8_in = (s64 *) &ptr_in[-offset]; //read in the aligned pointers
+ *
+ * data70 = *ptr8_in++;
+ * dataF8 = *ptr8_in++;
+ *
+ * data0 = HEXAGON_P_valignb_PPp(dataF8, data70, offset);
+ *
+ * prolog = 32 - ((int) ptr_out);
+ * mask = 0x7fffffff >> HEXAGON_R_cl0_R(len);
+ * prolog = prolog & mask;
+ * kernel = len - prolog;
+ * epilog = kernel & 0x1F;
+ * kernel = kernel>>5;
+ *
+ * if (prolog & 1) { ptr_out[0] = (u8) data0; data0 >>= 8; ptr_out += 1;}
+ * ptr2 = (s16 *) &ptr_out[0];
+ * if (prolog & 2) { ptr2[0] = (u16) data0; data0 >>= 16; ptr_out += 2;}
+ * ptr4 = (s32 *) &ptr_out[0];
+ * if (prolog & 4) { ptr4[0] = (u32) data0; data0 >>= 32; ptr_out += 4;}
+ *
+ * offset = offset + (prolog & 7);
+ * if (offset >= 8) {
+ * data70 = dataF8;
+ * dataF8 = *ptr8_in++;
+ * }
+ * offset = offset & 0x7;
+ *
+ * prolog = prolog >> 3;
+ * if (prolog) for (i=0; i < prolog; i++) {
+ * data0 = HEXAGON_P_valignb_PPp(dataF8, data70, offset);
+ * ptr8_out = (s64 *) &ptr_out[0]; *ptr8_out = data0; ptr_out += 8;
+ * data70 = dataF8;
+ * dataF8 = *ptr8_in++;
+ * }
+ * if(kernel) { kernel -= 1; epilog += 32; }
+ * if(kernel) for(i=0; i < kernel; i++) {
+ * data0 = HEXAGON_P_valignb_PPp(dataF8, data70, offset);
+ * ptr8_out = (s64 *) &ptr_out[0]; *ptr8_out = data0; ptr_out += 8;
+ * data70 = *ptr8_in++;
+ *
+ * data0 = HEXAGON_P_valignb_PPp(data70, dataF8, offset);
+ * ptr8_out = (s64 *) &ptr_out[0]; *ptr8_out = data0; ptr_out += 8;
+ * dataF8 = *ptr8_in++;
+ *
+ * data0 = HEXAGON_P_valignb_PPp(dataF8, data70, offset);
+ * ptr8_out = (s64 *) &ptr_out[0]; *ptr8_out = data0; ptr_out += 8;
+ * data70 = *ptr8_in++;
+ *
+ * data0 = HEXAGON_P_valignb_PPp(data70, dataF8, offset);
+ * ptr8_out = (s64 *) &ptr_out[0]; *ptr8_out = data0; ptr_out += 8;
+ * dataF8 = *ptr8_in++;
+ * }
+ * epilogdws = epilog >> 3;
+ * if (epilogdws) for (i=0; i < epilogdws; i++) {
+ * data0 = HEXAGON_P_valignb_PPp(dataF8, data70, offset);
+ * ptr8_out = (s64 *) &ptr_out[0]; *ptr8_out = data0; ptr_out += 8;
+ * data70 = dataF8;
+ * dataF8 = *ptr8_in++;
+ * }
+ * data0 = HEXAGON_P_valignb_PPp(dataF8, data70, offset);
+ *
+ * ptr4 = (s32 *) &ptr_out[0];
+ * if (epilog & 4) { ptr4[0] = (u32) data0; data0 >>= 32; ptr_out += 4;}
+ * ptr2 = (s16 *) &ptr_out[0];
+ * if (epilog & 2) { ptr2[0] = (u16) data0; data0 >>= 16; ptr_out += 2;}
+ * if (epilog & 1) { *ptr_out++ = (u8) data0; }
+ *
+ * return(ptr_out - length);
+ * }
+ *
+ * Codesize : 784 bytes
+ */
+
+
+#define ptr_out R0 /* destination pounter */
+#define ptr_in R1 /* source pointer */
+#define len R2 /* length of copy in bytes */
+
+#define data70 R13:12 /* lo 8 bytes of non-aligned transfer */
+#define dataF8 R11:10 /* hi 8 bytes of non-aligned transfer */
+#define ldata0 R7:6 /* even 8 bytes chunks */
+#define ldata1 R25:24 /* odd 8 bytes chunks */
+#define data1 R7 /* lower 8 bytes of ldata1 */
+#define data0 R6 /* lower 8 bytes of ldata0 */
+
+#define ifbyte p0 /* if transfer has bytes in epilog/prolog */
+#define ifhword p0 /* if transfer has shorts in epilog/prolog */
+#define ifword p0 /* if transfer has words in epilog/prolog */
+#define noprolog p0 /* no prolog, xfer starts at 32byte */
+#define nokernel p1 /* no 32byte multiple block in the transfer */
+#define noepilog p0 /* no epilog, xfer ends on 32byte boundary */
+#define align p2 /* alignment of input rel to 8byte boundary */
+#define kernel1 p0 /* kernel count == 1 */
+
+#define dalign R25 /* rel alignment of input to output data */
+#define star3 R16 /* number bytes in prolog - dwords */
+#define rest R8 /* length - prolog bytes */
+#define back R7 /* nr bytes > dword boundary in src block */
+#define epilog R3 /* bytes in epilog */
+#define inc R15:14 /* inc kernel by -1 and defetch ptr by 32 */
+#define kernel R4 /* number of 32byte chunks in kernel */
+#define ptr_in_p_128 R5 /* pointer for prefetch of input data */
+#define mask R8 /* mask used to determine prolog size */
+#define shift R8 /* used to work a shifter to extract bytes */
+#define shift2 R5 /* in epilog to workshifter to extract bytes */
+#define prolog R15 /* bytes in prolog */
+#define epilogdws R15 /* number dwords in epilog */
+#define shiftb R14 /* used to extract bytes */
+#define offset R9 /* same as align in reg */
+#define ptr_out_p_32 R17 /* pointer to output dczero */
+#define align888 R14 /* if simple dword loop can be used */
+#define len8 R9 /* number of dwords in length */
+#define over R20 /* nr of bytes > last inp buf dword boundary */
+
+#define ptr_in_p_128kernel R5:4 /* packed fetch pointer & kernel cnt */
+
+ .section .text
+ .p2align 4
+ .global memcpy
+ .type memcpy, @function
+memcpy:
+{
+ p2 = cmp.eq(len, #0); /* =0 */
+ align888 = or(ptr_in, ptr_out); /* %8 < 97 */
+ p0 = cmp.gtu(len, #23); /* %1, <24 */
+ p1 = cmp.eq(ptr_in, ptr_out); /* attempt to overwrite self */
+}
+{
+ p1 = or(p2, p1);
+ p3 = cmp.gtu(len, #95); /* %8 < 97 */
+ align888 = or(align888, len); /* %8 < 97 */
+ len8 = lsr(len, #3); /* %8 < 97 */
+}
+{
+ dcfetch(ptr_in); /* zero/ptrin=ptrout causes fetch */
+ p2 = bitsclr(align888, #7); /* %8 < 97 */
+ if(p1) jumpr r31; /* =0 */
+}
+{
+ p2 = and(p2,!p3); /* %8 < 97 */
+ if (p2.new) len = add(len, #-8); /* %8 < 97 */
+ if (p2.new) jump:NT .Ldwordaligned; /* %8 < 97 */
+}
+{
+ if(!p0) jump .Lbytes23orless; /* %1, <24 */
+ mask.l = #LO(0x7fffffff);
+ /* all bytes before line multiples of data */
+ prolog = sub(#0, ptr_out);
+}
+{
+ /* save r31 on stack, decrement sp by 16 */
+ allocframe(#24);
+ mask.h = #HI(0x7fffffff);
+ ptr_in_p_128 = add(ptr_in, #32);
+ back = cl0(len);
+}
+{
+ memd(sp+#0) = R17:16; /* save r16,r17 on stack6 */
+ r31.l = #LO(.Lmemcpy_return); /* set up final return pointer */
+ prolog &= lsr(mask, back);
+ offset = and(ptr_in, #7);
+}
+{
+ memd(sp+#8) = R25:24; /* save r25,r24 on stack */
+ dalign = sub(ptr_out, ptr_in);
+ r31.h = #HI(.Lmemcpy_return); /* set up final return pointer */
+}
+{
+ /* see if there if input buffer end if aligned */
+ over = add(len, ptr_in);
+ back = add(len, offset);
+ memd(sp+#16) = R21:20; /* save r20,r21 on stack */
+}
+{
+ noprolog = bitsclr(prolog, #7);
+ prolog = and(prolog, #31);
+ dcfetch(ptr_in_p_128);
+ ptr_in_p_128 = add(ptr_in_p_128, #32);
+}
+{
+ kernel = sub(len, prolog);
+ shift = asl(prolog, #3);
+ star3 = and(prolog, #7);
+ ptr_in = and(ptr_in, #-8);
+}
+{
+ prolog = lsr(prolog, #3);
+ epilog = and(kernel, #31);
+ ptr_out_p_32 = add(ptr_out, prolog);
+ over = and(over, #7);
+}
+{
+ p3 = cmp.gtu(back, #8);
+ kernel = lsr(kernel, #5);
+ dcfetch(ptr_in_p_128);
+ ptr_in_p_128 = add(ptr_in_p_128, #32);
+}
+{
+ p1 = cmp.eq(prolog, #0);
+ if(!p1.new) prolog = add(prolog, #1);
+ dcfetch(ptr_in_p_128); /* reserve the line 64bytes on */
+ ptr_in_p_128 = add(ptr_in_p_128, #32);
+}
+{
+ nokernel = cmp.eq(kernel,#0);
+ dcfetch(ptr_in_p_128); /* reserve the line 64bytes on */
+ ptr_in_p_128 = add(ptr_in_p_128, #32);
+ shiftb = and(shift, #8);
+}
+{
+ dcfetch(ptr_in_p_128); /* reserve the line 64bytes on */
+ ptr_in_p_128 = add(ptr_in_p_128, #32);
+ if(nokernel) jump .Lskip64;
+ p2 = cmp.eq(kernel, #1); /* skip ovr if kernel == 0 */
+}
+{
+ dczeroa(ptr_out_p_32);
+ /* don't advance pointer */
+ if(!p2) ptr_out_p_32 = add(ptr_out_p_32, #32);
+}
+{
+ dalign = and(dalign, #31);
+ dczeroa(ptr_out_p_32);
+}
+.Lskip64:
+{
+ data70 = memd(ptr_in++#16);
+ if(p3) dataF8 = memd(ptr_in+#8);
+ if(noprolog) jump .Lnoprolog32;
+ align = offset;
+}
+/* upto initial 7 bytes */
+{
+ ldata0 = valignb(dataF8, data70, align);
+ ifbyte = tstbit(shift,#3);
+ offset = add(offset, star3);
+}
+{
+ if(ifbyte) memb(ptr_out++#1) = data0;
+ ldata0 = lsr(ldata0, shiftb);
+ shiftb = and(shift, #16);
+ ifhword = tstbit(shift,#4);
+}
+{
+ if(ifhword) memh(ptr_out++#2) = data0;
+ ldata0 = lsr(ldata0, shiftb);
+ ifword = tstbit(shift,#5);
+ p2 = cmp.gtu(offset, #7);
+}
+{
+ if(ifword) memw(ptr_out++#4) = data0;
+ if(p2) data70 = dataF8;
+ if(p2) dataF8 = memd(ptr_in++#8); /* another 8 bytes */
+ align = offset;
+}
+.Lnoprolog32:
+{
+ p3 = sp1loop0(.Ldword_loop_prolog, prolog)
+ rest = sub(len, star3); /* whats left after the loop */
+ p0 = cmp.gt(over, #0);
+}
+ if(p0) rest = add(rest, #16);
+.Ldword_loop_prolog:
+{
+ if(p3) memd(ptr_out++#8) = ldata0;
+ ldata0 = valignb(dataF8, data70, align);
+ p0 = cmp.gt(rest, #16);
+}
+{
+ data70 = dataF8;
+ if(p0) dataF8 = memd(ptr_in++#8);
+ rest = add(rest, #-8);
+}:endloop0
+.Lkernel:
+{
+ /* kernel is at least 32bytes */
+ p3 = cmp.gtu(kernel, #0);
+ /* last itn. remove edge effects */
+ if(p3.new) kernel = add(kernel, #-1);
+ /* dealt with in last dword loop */
+ if(p3.new) epilog = add(epilog, #32);
+}
+{
+ nokernel = cmp.eq(kernel, #0); /* after adjustment, recheck */
+ if(nokernel.new) jump:NT .Lepilog; /* likely not taken */
+ inc = combine(#32, #-1);
+ p3 = cmp.gtu(dalign, #24);
+}
+{
+ if(p3) jump .Lodd_alignment;
+}
+{
+ loop0(.Loword_loop_25to31, kernel);
+ kernel1 = cmp.gtu(kernel, #1);
+ rest = kernel;
+}
+ .falign
+.Loword_loop_25to31:
+{
+ dcfetch(ptr_in_p_128); /* prefetch 4 lines ahead */
+ if(kernel1) ptr_out_p_32 = add(ptr_out_p_32, #32);
+}
+{
+ dczeroa(ptr_out_p_32); /* reserve the next 32bytes in cache */
+ p3 = cmp.eq(kernel, rest);
+}
+{
+ /* kernel -= 1 */
+ ptr_in_p_128kernel = vaddw(ptr_in_p_128kernel, inc);
+ /* kill write on first iteration */
+ if(!p3) memd(ptr_out++#8) = ldata1;
+ ldata1 = valignb(dataF8, data70, align);
+ data70 = memd(ptr_in++#8);
+}
+{
+ memd(ptr_out++#8) = ldata0;
+ ldata0 = valignb(data70, dataF8, align);
+ dataF8 = memd(ptr_in++#8);
+}
+{
+ memd(ptr_out++#8) = ldata1;
+ ldata1 = valignb(dataF8, data70, align);
+ data70 = memd(ptr_in++#8);
+}
+{
+ memd(ptr_out++#8) = ldata0;
+ ldata0 = valignb(data70, dataF8, align);
+ dataF8 = memd(ptr_in++#8);
+ kernel1 = cmp.gtu(kernel, #1);
+}:endloop0
+{
+ memd(ptr_out++#8) = ldata1;
+ jump .Lepilog;
+}
+.Lodd_alignment:
+{
+ loop0(.Loword_loop_00to24, kernel);
+ kernel1 = cmp.gtu(kernel, #1);
+ rest = add(kernel, #-1);
+}
+ .falign
+.Loword_loop_00to24:
+{
+ dcfetch(ptr_in_p_128); /* prefetch 4 lines ahead */
+ ptr_in_p_128kernel = vaddw(ptr_in_p_128kernel, inc);
+ if(kernel1) ptr_out_p_32 = add(ptr_out_p_32, #32);
+}
+{
+ dczeroa(ptr_out_p_32); /* reserve the next 32bytes in cache */
+}
+{
+ memd(ptr_out++#8) = ldata0;
+ ldata0 = valignb(dataF8, data70, align);
+ data70 = memd(ptr_in++#8);
+}
+{
+ memd(ptr_out++#8) = ldata0;
+ ldata0 = valignb(data70, dataF8, align);
+ dataF8 = memd(ptr_in++#8);
+}
+{
+ memd(ptr_out++#8) = ldata0;
+ ldata0 = valignb(dataF8, data70, align);
+ data70 = memd(ptr_in++#8);
+}
+{
+ memd(ptr_out++#8) = ldata0;
+ ldata0 = valignb(data70, dataF8, align);
+ dataF8 = memd(ptr_in++#8);
+ kernel1 = cmp.gtu(kernel, #1);
+}:endloop0
+.Lepilog:
+{
+ noepilog = cmp.eq(epilog,#0);
+ epilogdws = lsr(epilog, #3);
+ kernel = and(epilog, #7);
+}
+{
+ if(noepilog) jumpr r31;
+ if(noepilog) ptr_out = sub(ptr_out, len);
+ p3 = cmp.eq(epilogdws, #0);
+ shift2 = asl(epilog, #3);
+}
+{
+ shiftb = and(shift2, #32);
+ ifword = tstbit(epilog,#2);
+ if(p3) jump .Lepilog60;
+ if(!p3) epilog = add(epilog, #-16);
+}
+{
+ loop0(.Ldword_loop_epilog, epilogdws);
+ /* stop criteria is lsbs unless = 0 then its 8 */
+ p3 = cmp.eq(kernel, #0);
+ if(p3.new) kernel= #8;
+ p1 = cmp.gt(over, #0);
+}
+ /* if not aligned to end of buffer execute 1 more iteration */
+ if(p1) kernel= #0;
+.Ldword_loop_epilog:
+{
+ memd(ptr_out++#8) = ldata0;
+ ldata0 = valignb(dataF8, data70, align);
+ p3 = cmp.gt(epilog, kernel);
+}
+{
+ data70 = dataF8;
+ if(p3) dataF8 = memd(ptr_in++#8);
+ epilog = add(epilog, #-8);
+}:endloop0
+/* copy last 7 bytes */
+.Lepilog60:
+{
+ if(ifword) memw(ptr_out++#4) = data0;
+ ldata0 = lsr(ldata0, shiftb);
+ ifhword = tstbit(epilog,#1);
+ shiftb = and(shift2, #16);
+}
+{
+ if(ifhword) memh(ptr_out++#2) = data0;
+ ldata0 = lsr(ldata0, shiftb);
+ ifbyte = tstbit(epilog,#0);
+ if(ifbyte.new) len = add(len, #-1);
+}
+{
+ if(ifbyte) memb(ptr_out) = data0;
+ ptr_out = sub(ptr_out, len); /* return dest pointer */
+ jumpr r31;
+}
+/* do byte copy for small n */
+.Lbytes23orless:
+{
+ p3 = sp1loop0(.Lbyte_copy, len);
+ len = add(len, #-1);
+}
+.Lbyte_copy:
+{
+ data0 = memb(ptr_in++#1);
+ if(p3) memb(ptr_out++#1) = data0;
+}:endloop0
+{
+ memb(ptr_out) = data0;
+ ptr_out = sub(ptr_out, len);
+ jumpr r31;
+}
+/* do dword copies for aligned in, out and length */
+.Ldwordaligned:
+{
+ p3 = sp1loop0(.Ldword_copy, len8);
+}
+.Ldword_copy:
+{
+ if(p3) memd(ptr_out++#8) = ldata0;
+ ldata0 = memd(ptr_in++#8);
+}:endloop0
+{
+ memd(ptr_out) = ldata0;
+ ptr_out = sub(ptr_out, len);
+ jumpr r31; /* return to function caller */
+}
+.Lmemcpy_return:
+ r21:20 = memd(sp+#16); /* restore r20+r21 */
+{
+ r25:24 = memd(sp+#8); /* restore r24+r25 */
+ r17:16 = memd(sp+#0); /* restore r16+r17 */
+}
+ deallocframe; /* restore r31 and incrment stack by 16 */
+ jumpr r31
diff --git a/arch/hexagon/lib/memset.S b/arch/hexagon/lib/memset.S
new file mode 100644
index 0000000..e8a8f93
--- /dev/null
+++ b/arch/hexagon/lib/memset.S
@@ -0,0 +1,316 @@
+/*
+ * Copyright (c) 2011 Code Aurora Forum. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+
+/* HEXAGON assembly optimized memset */
+/* Replaces the standard library function memset */
+
+
+ .macro HEXAGON_OPT_FUNC_BEGIN name
+ .text
+ .p2align 4
+ .globl \name
+ .type \name, @function
+\name:
+ .endm
+
+ .macro HEXAGON_OPT_FUNC_FINISH name
+ .size \name, . - \name
+ .endm
+
+/* FUNCTION: memset (v2 version) */
+#if __HEXAGON_ARCH__ < 3
+HEXAGON_OPT_FUNC_BEGIN memset
+ {
+ r6 = #8
+ r7 = extractu(r0, #3 , #0)
+ p0 = cmp.eq(r2, #0)
+ p1 = cmp.gtu(r2, #7)
+ }
+ {
+ r4 = vsplatb(r1)
+ r8 = r0 /* leave r0 intact for return val */
+ r9 = sub(r6, r7) /* bytes until double alignment */
+ if p0 jumpr r31 /* count == 0, so return */
+ }
+ {
+ r3 = #0
+ r7 = #0
+ p0 = tstbit(r9, #0)
+ if p1 jump 2f /* skip byte loop */
+ }
+
+/* less than 8 bytes to set, so just set a byte at a time and return */
+
+ loop0(1f, r2) /* byte loop */
+ .falign
+1: /* byte loop */
+ {
+ memb(r8++#1) = r4
+ }:endloop0
+ jumpr r31
+ .falign
+2: /* skip byte loop */
+ {
+ r6 = #1
+ p0 = tstbit(r9, #1)
+ p1 = cmp.eq(r2, #1)
+ if !p0 jump 3f /* skip initial byte store */
+ }
+ {
+ memb(r8++#1) = r4
+ r3:2 = sub(r3:2, r7:6)
+ if p1 jumpr r31
+ }
+ .falign
+3: /* skip initial byte store */
+ {
+ r6 = #2
+ p0 = tstbit(r9, #2)
+ p1 = cmp.eq(r2, #2)
+ if !p0 jump 4f /* skip initial half store */
+ }
+ {
+ memh(r8++#2) = r4
+ r3:2 = sub(r3:2, r7:6)
+ if p1 jumpr r31
+ }
+ .falign
+4: /* skip initial half store */
+ {
+ r6 = #4
+ p0 = cmp.gtu(r2, #7)
+ p1 = cmp.eq(r2, #4)
+ if !p0 jump 5f /* skip initial word store */
+ }
+ {
+ memw(r8++#4) = r4
+ r3:2 = sub(r3:2, r7:6)
+ p0 = cmp.gtu(r2, #11)
+ if p1 jumpr r31
+ }
+ .falign
+5: /* skip initial word store */
+ {
+ r10 = lsr(r2, #3)
+ p1 = cmp.eq(r3, #1)
+ if !p0 jump 7f /* skip double loop */
+ }
+ {
+ r5 = r4
+ r6 = #8
+ loop0(6f, r10) /* double loop */
+ }
+
+/* set bytes a double word at a time */
+
+ .falign
+6: /* double loop */
+ {
+ memd(r8++#8) = r5:4
+ r3:2 = sub(r3:2, r7:6)
+ p1 = cmp.eq(r2, #8)
+ }:endloop0
+ .falign
+7: /* skip double loop */
+ {
+ p0 = tstbit(r2, #2)
+ if p1 jumpr r31
+ }
+ {
+ r6 = #4
+ p0 = tstbit(r2, #1)
+ p1 = cmp.eq(r2, #4)
+ if !p0 jump 8f /* skip final word store */
+ }
+ {
+ memw(r8++#4) = r4
+ r3:2 = sub(r3:2, r7:6)
+ if p1 jumpr r31
+ }
+ .falign
+8: /* skip final word store */
+ {
+ p1 = cmp.eq(r2, #2)
+ if !p0 jump 9f /* skip final half store */
+ }
+ {
+ memh(r8++#2) = r4
+ if p1 jumpr r31
+ }
+ .falign
+9: /* skip final half store */
+ {
+ memb(r8++#1) = r4
+ jumpr r31
+ }
+HEXAGON_OPT_FUNC_FINISH memset
+#endif
+
+
+/* FUNCTION: memset (v3 and higher version) */
+#if __HEXAGON_ARCH__ >= 3
+HEXAGON_OPT_FUNC_BEGIN memset
+ {
+ r7=vsplatb(r1)
+ r6 = r0
+ if (r2==#0) jump:nt .L1
+ }
+ {
+ r5:4=combine(r7,r7)
+ p0 = cmp.gtu(r2,#8)
+ if (p0.new) jump:nt .L3
+ }
+ {
+ r3 = r0
+ loop0(.L47,r2)
+ }
+ .falign
+.L47:
+ {
+ memb(r3++#1) = r1
+ }:endloop0 /* start=.L47 */
+ jumpr r31
+.L3:
+ {
+ p0 = tstbit(r0,#0)
+ if (!p0.new) jump:nt .L8
+ p1 = cmp.eq(r2, #1)
+ }
+ {
+ r6 = add(r0, #1)
+ r2 = add(r2,#-1)
+ memb(r0) = r1
+ if (p1) jump .L1
+ }
+.L8:
+ {
+ p0 = tstbit(r6,#1)
+ if (!p0.new) jump:nt .L10
+ }
+ {
+ r2 = add(r2,#-2)
+ memh(r6++#2) = r7
+ p0 = cmp.eq(r2, #2)
+ if (p0.new) jump:nt .L1
+ }
+.L10:
+ {
+ p0 = tstbit(r6,#2)
+ if (!p0.new) jump:nt .L12
+ }
+ {
+ r2 = add(r2,#-4)
+ memw(r6++#4) = r7
+ p0 = cmp.eq(r2, #4)
+ if (p0.new) jump:nt .L1
+ }
+.L12:
+ {
+ p0 = cmp.gtu(r2,#127)
+ if (!p0.new) jump:nt .L14
+ }
+ r3 = and(r6,#31)
+ if (r3==#0) jump:nt .L17
+ {
+ memd(r6++#8) = r5:4
+ r2 = add(r2,#-8)
+ }
+ r3 = and(r6,#31)
+ if (r3==#0) jump:nt .L17
+ {
+ memd(r6++#8) = r5:4
+ r2 = add(r2,#-8)
+ }
+ r3 = and(r6,#31)
+ if (r3==#0) jump:nt .L17
+ {
+ memd(r6++#8) = r5:4
+ r2 = add(r2,#-8)
+ }
+.L17:
+ {
+ r3 = lsr(r2,#5)
+ if (r1!=#0) jump:nt .L18
+ }
+ {
+ r8 = r3
+ r3 = r6
+ loop0(.L46,r3)
+ }
+ .falign
+.L46:
+ {
+ dczeroa(r6)
+ r6 = add(r6,#32)
+ r2 = add(r2,#-32)
+ }:endloop0 /* start=.L46 */
+.L14:
+ {
+ p0 = cmp.gtu(r2,#7)
+ if (!p0.new) jump:nt .L28
+ r8 = lsr(r2,#3)
+ }
+ loop0(.L44,r8)
+ .falign
+.L44:
+ {
+ memd(r6++#8) = r5:4
+ r2 = add(r2,#-8)
+ }:endloop0 /* start=.L44 */
+.L28:
+ {
+ p0 = tstbit(r2,#2)
+ if (!p0.new) jump:nt .L33
+ }
+ {
+ r2 = add(r2,#-4)
+ memw(r6++#4) = r7
+ }
+.L33:
+ {
+ p0 = tstbit(r2,#1)
+ if (!p0.new) jump:nt .L35
+ }
+ {
+ r2 = add(r2,#-2)
+ memh(r6++#2) = r7
+ }
+.L35:
+ p0 = cmp.eq(r2,#1)
+ if (p0) memb(r6) = r1
+.L1:
+ jumpr r31
+.L18:
+ loop0(.L45,r3)
+ .falign
+.L45:
+ dczeroa(r6)
+ {
+ memd(r6++#8) = r5:4
+ r2 = add(r2,#-32)
+ }
+ memd(r6++#8) = r5:4
+ memd(r6++#8) = r5:4
+ {
+ memd(r6++#8) = r5:4
+ }:endloop0 /* start=.L45 */
+ jump .L14
+HEXAGON_OPT_FUNC_FINISH memset
+#endif
+
--
1.7.1
--
Sent by an employee of the Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum.
^ permalink raw reply related
* [PATCH v5 09/36] Hexagon: Add checksum functions
From: Richard Kuo @ 2011-10-31 23:38 UTC (permalink / raw)
To: torvalds, linux-kernel
Signed-off-by: Richard Kuo <rkuo@codeaurora.org>
Signed-off-by: Linas Vepstas <linas@codeaurora.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
---
arch/hexagon/include/asm/checksum.h | 50 ++++++++
arch/hexagon/include/asm/intrinsics.h | 26 ++++
arch/hexagon/lib/checksum.c | 203 +++++++++++++++++++++++++++++++++
3 files changed, 279 insertions(+), 0 deletions(-)
create mode 100644 arch/hexagon/include/asm/checksum.h
create mode 100644 arch/hexagon/include/asm/intrinsics.h
create mode 100644 arch/hexagon/lib/checksum.c
diff --git a/arch/hexagon/include/asm/checksum.h b/arch/hexagon/include/asm/checksum.h
new file mode 100644
index 0000000..3ce4ecd
--- /dev/null
+++ b/arch/hexagon/include/asm/checksum.h
@@ -0,0 +1,50 @@
+/*
+ * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+#ifndef _ASM_CHECKSUM_H
+#define _ASM_CHECKSUM_H
+
+#define do_csum do_csum
+unsigned int do_csum(const void *voidptr, int len);
+
+/*
+ * the same as csum_partial, but copies from src while it
+ * checksums
+ *
+ * here even more important to align src and dst on a 32-bit (or even
+ * better 64-bit) boundary
+ */
+#define csum_partial_copy_nocheck csum_partial_copy_nocheck
+__wsum csum_partial_copy_nocheck(const void *src, void *dst,
+ int len, __wsum sum);
+
+/*
+ * computes the checksum of the TCP/UDP pseudo-header
+ * returns a 16-bit checksum, already complemented
+ */
+#define csum_tcpudp_nofold csum_tcpudp_nofold
+__wsum csum_tcpudp_nofold(unsigned long saddr, unsigned long daddr,
+ unsigned short len, unsigned short proto, __wsum sum);
+
+#define csum_tcpudp_magic csum_tcpudp_magic
+__sum16 csum_tcpudp_magic(unsigned long saddr, unsigned long daddr,
+ unsigned short len, unsigned short proto, __wsum sum);
+
+#include <asm-generic/checksum.h>
+
+#endif
diff --git a/arch/hexagon/include/asm/intrinsics.h b/arch/hexagon/include/asm/intrinsics.h
new file mode 100644
index 0000000..1c02186
--- /dev/null
+++ b/arch/hexagon/include/asm/intrinsics.h
@@ -0,0 +1,26 @@
+/*
+ * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+#ifndef _ASM_HEXAGON_INTRINSICS_H
+#define _ASM_HEXAGON_INTRINSICS_H
+
+#define HEXAGON_P_vrmpyhacc_PP __builtin_HEXAGON_M2_vrmac_s0
+#define HEXAGON_P_vrmpyh_PP __builtin_HEXAGON_M2_vrmpy_s0
+#define HEXAGON_R_cl0_R __builtin_HEXAGON_S2_cl0
+
+#endif
diff --git a/arch/hexagon/lib/checksum.c b/arch/hexagon/lib/checksum.c
new file mode 100644
index 0000000..9300552
--- /dev/null
+++ b/arch/hexagon/lib/checksum.c
@@ -0,0 +1,203 @@
+/*
+ * Checksum functions for Hexagon
+ *
+ * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+/* This was derived from arch/alpha/lib/checksum.c */
+
+
+#include <linux/module.h>
+#include <linux/string.h>
+
+#include <asm/byteorder.h>
+#include <net/checksum.h>
+#include <linux/uaccess.h>
+#include <asm/intrinsics.h>
+
+
+/* Vector value operations */
+#define SIGN(x, y) ((0x8000ULL*x)<<y)
+#define CARRY(x, y) ((0x0002ULL*x)<<y)
+#define SELECT(x, y) ((0x0001ULL*x)<<y)
+
+#define VR_NEGATE(a, b, c, d) (SIGN(a, 48) + SIGN(b, 32) + SIGN(c, 16) \
+ + SIGN(d, 0))
+#define VR_CARRY(a, b, c, d) (CARRY(a, 48) + CARRY(b, 32) + CARRY(c, 16) \
+ + CARRY(d, 0))
+#define VR_SELECT(a, b, c, d) (SELECT(a, 48) + SELECT(b, 32) + SELECT(c, 16) \
+ + SELECT(d, 0))
+
+
+/* optimized HEXAGON V3 intrinsic version */
+static inline unsigned short from64to16(u64 x)
+{
+ u64 sum;
+
+ sum = HEXAGON_P_vrmpyh_PP(x^VR_NEGATE(1, 1, 1, 1),
+ VR_SELECT(1, 1, 1, 1));
+ sum += VR_CARRY(0, 0, 1, 0);
+ sum = HEXAGON_P_vrmpyh_PP(sum, VR_SELECT(0, 0, 1, 1));
+
+ return 0xFFFF & sum;
+}
+
+/*
+ * computes the checksum of the TCP/UDP pseudo-header
+ * returns a 16-bit checksum, already complemented.
+ */
+__sum16 csum_tcpudp_magic(unsigned long saddr, unsigned long daddr,
+ unsigned short len, unsigned short proto,
+ __wsum sum)
+{
+ return (__force __sum16)~from64to16(
+ (__force u64)saddr + (__force u64)daddr +
+ (__force u64)sum + ((len + proto) << 8));
+}
+
+__wsum csum_tcpudp_nofold(unsigned long saddr, unsigned long daddr,
+ unsigned short len, unsigned short proto,
+ __wsum sum)
+{
+ u64 result;
+
+ result = (__force u64)saddr + (__force u64)daddr +
+ (__force u64)sum + ((len + proto) << 8);
+
+ /* Fold down to 32-bits so we don't lose in the typedef-less
+ network stack. */
+ /* 64 to 33 */
+ result = (result & 0xffffffffUL) + (result >> 32);
+ /* 33 to 32 */
+ result = (result & 0xffffffffUL) + (result >> 32);
+ return (__force __wsum)result;
+}
+EXPORT_SYMBOL(csum_tcpudp_nofold);
+
+/*
+ * Do a 64-bit checksum on an arbitrary memory area..
+ *
+ * This isn't a great routine, but it's not _horrible_ either. The
+ * inner loop could be unrolled a bit further, and there are better
+ * ways to do the carry, but this is reasonable.
+ */
+
+/* optimized HEXAGON intrinsic version, with over read fixed */
+unsigned int do_csum(const void *voidptr, int len)
+{
+ u64 sum0, sum1, x0, x1, *ptr8_o, *ptr8_e, *ptr8;
+ int i, start, mid, end, mask;
+ const char *ptr = voidptr;
+ unsigned short *ptr2;
+ unsigned int *ptr4;
+
+ if (len <= 0)
+ return 0;
+
+ start = 0xF & (16-(((int) ptr) & 0xF)) ;
+ mask = 0x7fffffffUL >> HEXAGON_R_cl0_R(len);
+ start = start & mask ;
+
+ mid = len - start;
+ end = mid & 0xF;
+ mid = mid>>4;
+ sum0 = mid << 18;
+ sum1 = 0;
+
+ if (start & 1)
+ sum0 += (u64) (ptr[0] << 8);
+ ptr2 = (unsigned short *) &ptr[start & 1];
+ if (start & 2)
+ sum1 += (u64) ptr2[0];
+ ptr4 = (unsigned int *) &ptr[start & 3];
+ if (start & 4) {
+ sum0 = HEXAGON_P_vrmpyhacc_PP(sum0,
+ VR_NEGATE(0, 0, 1, 1)^((u64)ptr4[0]),
+ VR_SELECT(0, 0, 1, 1));
+ sum0 += VR_SELECT(0, 0, 1, 0);
+ }
+ ptr8 = (u64 *) &ptr[start & 7];
+ if (start & 8) {
+ sum1 = HEXAGON_P_vrmpyhacc_PP(sum1,
+ VR_NEGATE(1, 1, 1, 1)^(ptr8[0]),
+ VR_SELECT(1, 1, 1, 1));
+ sum1 += VR_CARRY(0, 0, 1, 0);
+ }
+ ptr8_o = (u64 *) (ptr + start);
+ ptr8_e = (u64 *) (ptr + start + 8);
+
+ if (mid) {
+ x0 = *ptr8_e; ptr8_e += 2;
+ x1 = *ptr8_o; ptr8_o += 2;
+ if (mid > 1)
+ for (i = 0; i < mid-1; i++) {
+ sum0 = HEXAGON_P_vrmpyhacc_PP(sum0,
+ x0^VR_NEGATE(1, 1, 1, 1),
+ VR_SELECT(1, 1, 1, 1));
+ sum1 = HEXAGON_P_vrmpyhacc_PP(sum1,
+ x1^VR_NEGATE(1, 1, 1, 1),
+ VR_SELECT(1, 1, 1, 1));
+ x0 = *ptr8_e; ptr8_e += 2;
+ x1 = *ptr8_o; ptr8_o += 2;
+ }
+ sum0 = HEXAGON_P_vrmpyhacc_PP(sum0, x0^VR_NEGATE(1, 1, 1, 1),
+ VR_SELECT(1, 1, 1, 1));
+ sum1 = HEXAGON_P_vrmpyhacc_PP(sum1, x1^VR_NEGATE(1, 1, 1, 1),
+ VR_SELECT(1, 1, 1, 1));
+ }
+
+ ptr4 = (unsigned int *) &ptr[start + (mid * 16) + (end & 8)];
+ if (end & 4) {
+ sum1 = HEXAGON_P_vrmpyhacc_PP(sum1,
+ VR_NEGATE(0, 0, 1, 1)^((u64)ptr4[0]),
+ VR_SELECT(0, 0, 1, 1));
+ sum1 += VR_SELECT(0, 0, 1, 0);
+ }
+ ptr2 = (unsigned short *) &ptr[start + (mid * 16) + (end & 12)];
+ if (end & 2)
+ sum0 += (u64) ptr2[0];
+
+ if (end & 1)
+ sum1 += (u64) ptr[start + (mid * 16) + (end & 14)];
+
+ ptr8 = (u64 *) &ptr[start + (mid * 16)];
+ if (end & 8) {
+ sum0 = HEXAGON_P_vrmpyhacc_PP(sum0,
+ VR_NEGATE(1, 1, 1, 1)^(ptr8[0]),
+ VR_SELECT(1, 1, 1, 1));
+ sum0 += VR_CARRY(0, 0, 1, 0);
+ }
+ sum0 = HEXAGON_P_vrmpyh_PP((sum0+sum1)^VR_NEGATE(0, 0, 0, 1),
+ VR_SELECT(0, 0, 1, 1));
+ sum0 += VR_NEGATE(0, 0, 0, 1);
+ sum0 = HEXAGON_P_vrmpyh_PP(sum0, VR_SELECT(0, 0, 1, 1));
+
+ if (start & 1)
+ sum0 = (sum0 << 8) | (0xFF & (sum0 >> 8));
+
+ return 0xFFFF & sum0;
+}
+
+/*
+ * copy from ds while checksumming, otherwise like csum_partial
+ */
+__wsum
+csum_partial_copy_nocheck(const void *src, void *dst, int len, __wsum sum)
+{
+ memcpy(dst, src, len);
+ return csum_partial(dst, len, sum);
+}
--
1.7.1
--
Sent by an employee of the Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum.
^ permalink raw reply related
* [PATCH v5 06/36] Hexagon: Add processor and system headers
From: Richard Kuo @ 2011-10-31 23:36 UTC (permalink / raw)
To: torvalds, linux-kernel
Signed-off-by: Richard Kuo <rkuo@codeaurora.org>
Signed-off-by: Linas Vepstas <linas@codeaurora.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
---
arch/hexagon/include/asm/asm-offsets.h | 1 +
arch/hexagon/include/asm/irqflags.h | 62 +++++++++
arch/hexagon/include/asm/processor.h | 123 +++++++++++++++++
arch/hexagon/include/asm/registers.h | 236 ++++++++++++++++++++++++++++++++
arch/hexagon/include/asm/system.h | 126 +++++++++++++++++
5 files changed, 548 insertions(+), 0 deletions(-)
create mode 100644 arch/hexagon/include/asm/asm-offsets.h
create mode 100644 arch/hexagon/include/asm/irqflags.h
create mode 100644 arch/hexagon/include/asm/processor.h
create mode 100644 arch/hexagon/include/asm/registers.h
create mode 100644 arch/hexagon/include/asm/system.h
diff --git a/arch/hexagon/include/asm/asm-offsets.h b/arch/hexagon/include/asm/asm-offsets.h
new file mode 100644
index 0000000..d370ee3
--- /dev/null
+++ b/arch/hexagon/include/asm/asm-offsets.h
@@ -0,0 +1 @@
+#include <generated/asm-offsets.h>
diff --git a/arch/hexagon/include/asm/irqflags.h b/arch/hexagon/include/asm/irqflags.h
new file mode 100644
index 0000000..ec15236
--- /dev/null
+++ b/arch/hexagon/include/asm/irqflags.h
@@ -0,0 +1,62 @@
+/*
+ * IRQ support for the Hexagon architecture
+ *
+ * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+#ifndef _ASM_IRQFLAGS_H
+#define _ASM_IRQFLAGS_H
+
+#include <asm/hexagon_vm.h>
+#include <linux/types.h>
+
+static inline unsigned long arch_local_save_flags(void)
+{
+ return __vmgetie();
+}
+
+static inline unsigned long arch_local_irq_save(void)
+{
+ return __vmsetie(VM_INT_DISABLE);
+}
+
+static inline bool arch_irqs_disabled_flags(unsigned long flags)
+{
+ return !flags;
+}
+
+static inline bool arch_irqs_disabled(void)
+{
+ return !__vmgetie();
+}
+
+static inline void arch_local_irq_enable(void)
+{
+ __vmsetie(VM_INT_ENABLE);
+}
+
+static inline void arch_local_irq_disable(void)
+{
+ __vmsetie(VM_INT_DISABLE);
+}
+
+static inline void arch_local_irq_restore(unsigned long flags)
+{
+ __vmsetie(flags);
+}
+
+#endif
diff --git a/arch/hexagon/include/asm/processor.h b/arch/hexagon/include/asm/processor.h
new file mode 100644
index 0000000..20c5dda
--- /dev/null
+++ b/arch/hexagon/include/asm/processor.h
@@ -0,0 +1,123 @@
+/*
+ * Process/processor support for the Hexagon architecture
+ *
+ * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+#ifndef _ASM_PROCESSOR_H
+#define _ASM_PROCESSOR_H
+
+#ifndef __ASSEMBLY__
+
+#include <asm/mem-layout.h>
+#include <asm/registers.h>
+#include <asm/hexagon_vm.h>
+
+/* must be a macro */
+#define current_text_addr() ({ __label__ _l; _l: &&_l; })
+
+/* task_struct, defined elsewhere, is the "process descriptor" */
+struct task_struct;
+
+/* this is defined in arch/process.c */
+extern pid_t kernel_thread(int (*fn)(void *), void * arg, unsigned long flags);
+extern unsigned long thread_saved_pc(struct task_struct *tsk);
+
+extern void start_thread(struct pt_regs *, unsigned long, unsigned long);
+
+/*
+ * thread_struct is supposed to be for context switch data.
+ * Specifically, to hold the state necessary to perform switch_to...
+ */
+struct thread_struct {
+ void *switch_sp;
+};
+
+/*
+ * initializes thread_struct
+ * The only thing we have in there is switch_sp
+ * which doesn't really need to be initialized.
+ */
+
+#define INIT_THREAD { \
+}
+
+#define cpu_relax() __vmyield()
+
+/*
+ * "Unlazying all lazy status" occurs here.
+ */
+static inline void prepare_to_copy(struct task_struct *tsk)
+{
+}
+
+/*
+ * Decides where the kernel will search for a free chunk of vm space during
+ * mmaps.
+ * See also arch_get_unmapped_area.
+ * Doesn't affect if you have MAX_FIXED in the page flags set though...
+ *
+ * Apparently the convention is that ld.so will ask for "unmapped" private
+ * memory to be allocated SOMEWHERE, but it also asks for memory explicitly
+ * via MAP_FIXED at the lower * addresses starting at VA=0x0.
+ *
+ * If the two requests collide, you get authentic segfaulting action, so
+ * you have to kick the "unmapped" base requests higher up.
+ */
+#define TASK_UNMAPPED_BASE (PAGE_ALIGN(TASK_SIZE/3))
+
+
+#define task_pt_regs(task) \
+ ((struct pt_regs *)(task_stack_page(task) + THREAD_SIZE) - 1)
+
+#define KSTK_EIP(tsk) (pt_elr(task_pt_regs(tsk)))
+#define KSTK_ESP(tsk) (pt_psp(task_pt_regs(tsk)))
+
+/* Free all resources held by a thread; defined in process.c */
+extern void release_thread(struct task_struct *dead_task);
+
+/* Get wait channel for task P. */
+extern unsigned long get_wchan(struct task_struct *p);
+
+/* The following stuff is pretty HEXAGON specific. */
+
+/* This is really just here for __switch_to.
+ Offsets are pulled via asm-offsets.c */
+
+/*
+ * No real reason why VM and native switch stacks should be different.
+ * Ultimately this should merge. Note that Rev C. ABI called out only
+ * R24-27 as callee saved GPRs needing explicit attention (R29-31 being
+ * dealt with automagically by allocframe), but the current ABI has
+ * more, R16-R27. By saving more, the worst case is that we waste some
+ * cycles if building with the old compilers.
+ */
+
+struct hexagon_switch_stack {
+ unsigned long long r1716;
+ unsigned long long r1918;
+ unsigned long long r2120;
+ unsigned long long r2322;
+ unsigned long long r2524;
+ unsigned long long r2726;
+ unsigned long fp;
+ unsigned long lr;
+};
+
+#endif /* !__ASSEMBLY__ */
+
+#endif
diff --git a/arch/hexagon/include/asm/registers.h b/arch/hexagon/include/asm/registers.h
new file mode 100644
index 0000000..4dd741b
--- /dev/null
+++ b/arch/hexagon/include/asm/registers.h
@@ -0,0 +1,236 @@
+/*
+ * Register definitions for the Hexagon architecture
+ *
+ * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+#ifndef _ASM_REGISTERS_H
+#define _ASM_REGISTERS_H
+
+#define SP r29
+
+#ifndef __ASSEMBLY__
+
+/* See kernel/entry.S for further documentation. */
+
+/*
+ * Entry code copies the event record out of guest registers into
+ * this structure (which is on the stack).
+ */
+
+struct hvm_event_record {
+ unsigned long vmel; /* Event Linkage (return address) */
+ unsigned long vmest; /* Event context - pre-event SSR values */
+ unsigned long vmpsp; /* Previous stack pointer */
+ unsigned long vmbadva; /* Bad virtual address for addressing events */
+};
+
+struct pt_regs {
+ long restart_r0; /* R0 checkpoint for syscall restart */
+ long syscall_nr; /* Only used in system calls */
+ union {
+ struct {
+ unsigned long usr;
+ unsigned long preds;
+ };
+ long long int predsusr;
+ };
+ union {
+ struct {
+ unsigned long m0;
+ unsigned long m1;
+ };
+ long long int m1m0;
+ };
+ union {
+ struct {
+ unsigned long sa1;
+ unsigned long lc1;
+ };
+ long long int lc1sa1;
+ };
+ union {
+ struct {
+ unsigned long sa0;
+ unsigned long lc0;
+ };
+ long long int lc0sa0;
+ };
+ union {
+ struct {
+ unsigned long gp;
+ unsigned long ugp;
+ };
+ long long int ugpgp;
+ };
+ /*
+ * Be extremely careful with rearranging these, if at all. Some code
+ * assumes the 32 registers exist exactly like this in memory;
+ * e.g. kernel/ptrace.c
+ * e.g. kernel/signal.c (restore_sigcontext)
+ */
+ union {
+ struct {
+ unsigned long r00;
+ unsigned long r01;
+ };
+ long long int r0100;
+ };
+ union {
+ struct {
+ unsigned long r02;
+ unsigned long r03;
+ };
+ long long int r0302;
+ };
+ union {
+ struct {
+ unsigned long r04;
+ unsigned long r05;
+ };
+ long long int r0504;
+ };
+ union {
+ struct {
+ unsigned long r06;
+ unsigned long r07;
+ };
+ long long int r0706;
+ };
+ union {
+ struct {
+ unsigned long r08;
+ unsigned long r09;
+ };
+ long long int r0908;
+ };
+ union {
+ struct {
+ unsigned long r10;
+ unsigned long r11;
+ };
+ long long int r1110;
+ };
+ union {
+ struct {
+ unsigned long r12;
+ unsigned long r13;
+ };
+ long long int r1312;
+ };
+ union {
+ struct {
+ unsigned long r14;
+ unsigned long r15;
+ };
+ long long int r1514;
+ };
+ union {
+ struct {
+ unsigned long r16;
+ unsigned long r17;
+ };
+ long long int r1716;
+ };
+ union {
+ struct {
+ unsigned long r18;
+ unsigned long r19;
+ };
+ long long int r1918;
+ };
+ union {
+ struct {
+ unsigned long r20;
+ unsigned long r21;
+ };
+ long long int r2120;
+ };
+ union {
+ struct {
+ unsigned long r22;
+ unsigned long r23;
+ };
+ long long int r2322;
+ };
+ union {
+ struct {
+ unsigned long r24;
+ unsigned long r25;
+ };
+ long long int r2524;
+ };
+ union {
+ struct {
+ unsigned long r26;
+ unsigned long r27;
+ };
+ long long int r2726;
+ };
+ union {
+ struct {
+ unsigned long r28;
+ unsigned long r29;
+ };
+ long long int r2928;
+ };
+ union {
+ struct {
+ unsigned long r30;
+ unsigned long r31;
+ };
+ long long int r3130;
+ };
+ /* VM dispatch pushes event record onto stack - we can build on it */
+ struct hvm_event_record hvmer;
+};
+
+/* Defines to conveniently access the values */
+
+/*
+ * As of the VM spec 0.5, these registers are now set/retrieved via a
+ * VM call. On the in-bound side, we just fetch the values
+ * at the entry points and stuff them into the old record in pt_regs.
+ * However, on the outbound side, probably at VM rte, we set the
+ * registers back.
+ */
+
+#define pt_elr(regs) ((regs)->hvmer.vmel)
+#define pt_set_elr(regs, val) ((regs)->hvmer.vmel = (val))
+#define pt_cause(regs) ((regs)->hvmer.vmest & (HVM_VMEST_CAUSE_MSK))
+#define user_mode(regs) \
+ (((regs)->hvmer.vmest & (HVM_VMEST_UM_MSK << HVM_VMEST_UM_SFT)) != 0)
+#define ints_enabled(regs) \
+ (((regs)->hvmer.vmest & (HVM_VMEST_IE_MSK << HVM_VMEST_IE_SFT)) != 0)
+#define pt_psp(regs) ((regs)->hvmer.vmpsp)
+#define pt_badva(regs) ((regs)->hvmer.vmbadva)
+
+#define pt_set_rte_sp(regs, sp) do {\
+ pt_psp(regs) = (sp);\
+ (regs)->SP = (unsigned long) &((regs)->hvmer);\
+ } while (0)
+
+#define pt_set_kmode(regs) \
+ (regs)->hvmer.vmest = (HVM_VMEST_IE_MSK << HVM_VMEST_IE_SFT)
+
+#define pt_set_usermode(regs) \
+ (regs)->hvmer.vmest = (HVM_VMEST_UM_MSK << HVM_VMEST_UM_SFT) \
+ | (HVM_VMEST_IE_MSK << HVM_VMEST_IE_SFT)
+
+#endif /* ifndef __ASSEMBLY */
+
+#endif
diff --git a/arch/hexagon/include/asm/system.h b/arch/hexagon/include/asm/system.h
new file mode 100644
index 0000000..323ed1d
--- /dev/null
+++ b/arch/hexagon/include/asm/system.h
@@ -0,0 +1,126 @@
+/*
+ * System level definitions for the Hexagon architecture
+ *
+ * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+#ifndef _ASM_SYSTEM_H
+#define _ASM_SYSTEM_H
+
+#include <linux/linkage.h>
+#include <linux/irqflags.h>
+#include <asm/atomic.h>
+#include <asm/hexagon_vm.h>
+
+struct thread_struct;
+
+extern struct task_struct *__switch_to(struct task_struct *,
+ struct task_struct *,
+ struct task_struct *);
+
+#define switch_to(p, n, r) do {\
+ r = __switch_to((p), (n), (r));\
+} while (0)
+
+
+#define rmb() barrier()
+#define read_barrier_depends() barrier()
+#define wmb() barrier()
+#define mb() barrier()
+#define smp_rmb() barrier()
+#define smp_read_barrier_depends() barrier()
+#define smp_wmb() barrier()
+#define smp_mb() barrier()
+#define smp_mb__before_atomic_dec() barrier()
+#define smp_mb__after_atomic_dec() barrier()
+#define smp_mb__before_atomic_inc() barrier()
+#define smp_mb__after_atomic_inc() barrier()
+
+/*
+ * __xchg - atomically exchange a register and a memory location
+ * @x: value to swap
+ * @ptr: pointer to memory
+ * @size: size of the value
+ *
+ * Only 4 bytes supported currently.
+ *
+ * Note: there was an errata for V2 about .new's and memw_locked.
+ *
+ */
+static inline unsigned long __xchg(unsigned long x, volatile void *ptr,
+ int size)
+{
+ unsigned long retval;
+
+ /* Can't seem to use printk or panic here, so just stop */
+ if (size != 4) do { asm volatile("brkpt;\n"); } while (1);
+
+ __asm__ __volatile__ (
+ "1: %0 = memw_locked(%1);\n" /* load into retval */
+ " memw_locked(%1,P0) = %2;\n" /* store into memory */
+ " if !P0 jump 1b;\n"
+ : "=&r" (retval)
+ : "r" (ptr), "r" (x)
+ : "memory", "p0"
+ );
+ return retval;
+}
+
+/*
+ * Atomically swap the contents of a register with memory. Should be atomic
+ * between multiple CPU's and within interrupts on the same CPU.
+ */
+#define xchg(ptr, v) ((__typeof__(*(ptr)))__xchg((unsigned long)(v), (ptr), \
+ sizeof(*(ptr))))
+
+/* Set a value and use a memory barrier. Used by the scheduler somewhere. */
+#define set_mb(var, value) \
+ do { var = value; mb(); } while (0)
+
+/*
+ * see rt-mutex-design.txt; cmpxchg supposedly checks if *ptr == A and swaps.
+ * looks just like atomic_cmpxchg on our arch currently with a bunch of
+ * variable casting.
+ */
+#define __HAVE_ARCH_CMPXCHG 1
+
+#define cmpxchg(ptr, old, new) \
+({ \
+ __typeof__(ptr) __ptr = (ptr); \
+ __typeof__(*(ptr)) __old = (old); \
+ __typeof__(*(ptr)) __new = (new); \
+ __typeof__(*(ptr)) __oldval = 0; \
+ \
+ asm volatile( \
+ "1: %0 = memw_locked(%1);\n" \
+ " { P0 = cmp.eq(%0,%2);\n" \
+ " if (!P0.new) jump:nt 2f; }\n" \
+ " memw_locked(%1,p0) = %3;\n" \
+ " if (!P0) jump 1b;\n" \
+ "2:\n" \
+ : "=&r" (__oldval) \
+ : "r" (__ptr), "r" (__old), "r" (__new) \
+ : "memory", "p0" \
+ ); \
+ __oldval; \
+})
+
+/* Should probably shoot for an 8-byte aligned stack pointer */
+#define STACK_MASK (~7)
+#define arch_align_stack(x) (x & STACK_MASK)
+
+#endif
--
1.7.1
--
Sent by an employee of the Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum.
^ permalink raw reply related
* [PATCH v5 04/36] Hexagon: Add atomic ops support
From: Richard Kuo @ 2011-10-31 23:28 UTC (permalink / raw)
To: torvalds, linux-kernel
Signed-off-by: Richard Kuo <rkuo@codeaurora.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
---
arch/hexagon/include/asm/atomic.h | 164 +++++++++++++++++++++++++++++++++++++
1 files changed, 164 insertions(+), 0 deletions(-)
create mode 100644 arch/hexagon/include/asm/atomic.h
diff --git a/arch/hexagon/include/asm/atomic.h b/arch/hexagon/include/asm/atomic.h
new file mode 100644
index 0000000..e220f90
--- /dev/null
+++ b/arch/hexagon/include/asm/atomic.h
@@ -0,0 +1,164 @@
+/*
+ * Atomic operations for the Hexagon architecture
+ *
+ * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved.
+ *
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+#ifndef _ASM_ATOMIC_H
+#define _ASM_ATOMIC_H
+
+#include <linux/types.h>
+
+#define ATOMIC_INIT(i) { (i) }
+#define atomic_set(v, i) ((v)->counter = (i))
+
+/**
+ * atomic_read - reads a word, atomically
+ * @v: pointer to atomic value
+ *
+ * Assumes all word reads on our architecture are atomic.
+ */
+#define atomic_read(v) ((v)->counter)
+
+/**
+ * atomic_xchg - atomic
+ * @v: pointer to memory to change
+ * @new: new value (technically passed in a register -- see xchg)
+ */
+#define atomic_xchg(v, new) (xchg(&((v)->counter), (new)))
+
+
+/**
+ * atomic_cmpxchg - atomic compare-and-exchange values
+ * @v: pointer to value to change
+ * @old: desired old value to match
+ * @new: new value to put in
+ *
+ * Parameters are then pointer, value-in-register, value-in-register,
+ * and the output is the old value.
+ *
+ * Apparently this is complicated for archs that don't support
+ * the memw_locked like we do (or it's broken or whatever).
+ *
+ * Kind of the lynchpin of the rest of the generically defined routines.
+ * Remember V2 had that bug with dotnew predicate set by memw_locked.
+ *
+ * "old" is "expected" old val, __oldval is actual old value
+ */
+static inline int atomic_cmpxchg(atomic_t *v, int old, int new)
+{
+ int __oldval;
+
+ asm volatile(
+ "1: %0 = memw_locked(%1);\n"
+ " { P0 = cmp.eq(%0,%2);\n"
+ " if (!P0.new) jump:nt 2f; }\n"
+ " memw_locked(%1,P0) = %3;\n"
+ " if (!P0) jump 1b;\n"
+ "2:\n"
+ : "=&r" (__oldval)
+ : "r" (&v->counter), "r" (old), "r" (new)
+ : "memory", "p0"
+ );
+
+ return __oldval;
+}
+
+static inline int atomic_add_return(int i, atomic_t *v)
+{
+ int output;
+
+ __asm__ __volatile__ (
+ "1: %0 = memw_locked(%1);\n"
+ " %0 = add(%0,%2);\n"
+ " memw_locked(%1,P3)=%0;\n"
+ " if !P3 jump 1b;\n"
+ : "=&r" (output)
+ : "r" (&v->counter), "r" (i)
+ : "memory", "p3"
+ );
+ return output;
+
+}
+
+#define atomic_add(i, v) atomic_add_return(i, (v))
+
+static inline int atomic_sub_return(int i, atomic_t *v)
+{
+ int output;
+ __asm__ __volatile__ (
+ "1: %0 = memw_locked(%1);\n"
+ " %0 = sub(%0,%2);\n"
+ " memw_locked(%1,P3)=%0\n"
+ " if !P3 jump 1b;\n"
+ : "=&r" (output)
+ : "r" (&v->counter), "r" (i)
+ : "memory", "p3"
+ );
+ return output;
+}
+
+#define atomic_sub(i, v) atomic_sub_return(i, (v))
+
+/**
+ * atomic_add_unless - add unless the number is a given value
+ * @v: pointer to value
+ * @a: amount to add
+ * @u: unless value is equal to u
+ *
+ * Returns 1 if the add happened, 0 if it didn't.
+ */
+static inline int __atomic_add_unless(atomic_t *v, int a, int u)
+{
+ int output, __oldval;
+ asm volatile(
+ "1: %0 = memw_locked(%2);"
+ " {"
+ " p3 = cmp.eq(%0, %4);"
+ " if (p3.new) jump:nt 2f;"
+ " %0 = add(%0, %3);"
+ " %1 = #0;"
+ " }"
+ " memw_locked(%2, p3) = %0;"
+ " {"
+ " if !p3 jump 1b;"
+ " %1 = #1;"
+ " }"
+ "2:"
+ : "=&r" (__oldval), "=&r" (output)
+ : "r" (v), "r" (a), "r" (u)
+ : "memory", "p3"
+ );
+ return output;
+}
+
+#define atomic_inc_not_zero(v) atomic_add_unless((v), 1, 0)
+
+#define atomic_inc(v) atomic_add(1, (v))
+#define atomic_dec(v) atomic_sub(1, (v))
+
+#define atomic_inc_and_test(v) (atomic_add_return(1, (v)) == 0)
+#define atomic_dec_and_test(v) (atomic_sub_return(1, (v)) == 0)
+#define atomic_sub_and_test(i, v) (atomic_sub_return(i, (v)) == 0)
+#define atomic_add_negative(i, v) (atomic_add_return(i, (v)) < 0)
+
+
+#define atomic_inc_return(v) (atomic_add_return(1, v))
+#define atomic_dec_return(v) (atomic_sub_return(1, v))
+
+#endif
--
1.7.1
--
Sent by an employee of the Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum.
^ permalink raw reply related
* [PATCH v5 03/36] Hexagon: Add bitops support
From: Richard Kuo @ 2011-10-31 23:27 UTC (permalink / raw)
To: torvalds, linux-kernel
Signed-off-by: Richard Kuo <rkuo@codeaurora.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
---
arch/hexagon/include/asm/bitops.h | 301 +++++++++++++++++++++++++++++++++++++
1 files changed, 301 insertions(+), 0 deletions(-)
create mode 100644 arch/hexagon/include/asm/bitops.h
diff --git a/arch/hexagon/include/asm/bitops.h b/arch/hexagon/include/asm/bitops.h
new file mode 100644
index 0000000..d23461e
--- /dev/null
+++ b/arch/hexagon/include/asm/bitops.h
@@ -0,0 +1,301 @@
+/*
+ * Bit operations for the Hexagon architecture
+ *
+ * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved.
+ *
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+#ifndef _ASM_BITOPS_H
+#define _ASM_BITOPS_H
+
+#include <linux/compiler.h>
+#include <asm/byteorder.h>
+#include <asm/system.h>
+#include <asm/atomic.h>
+
+#ifdef __KERNEL__
+
+#define smp_mb__before_clear_bit() barrier()
+#define smp_mb__after_clear_bit() barrier()
+
+/*
+ * The offset calculations for these are based on BITS_PER_LONG == 32
+ * (i.e. I get to shift by #5-2 (32 bits per long, 4 bytes per access),
+ * mask by 0x0000001F)
+ *
+ * Typically, R10 is clobbered for address, R11 bit nr, and R12 is temp
+ */
+
+/**
+ * test_and_clear_bit - clear a bit and return its old value
+ * @nr: bit number to clear
+ * @addr: pointer to memory
+ */
+static inline int test_and_clear_bit(int nr, volatile void *addr)
+{
+ int oldval;
+
+ __asm__ __volatile__ (
+ " {R10 = %1; R11 = asr(%2,#5); }\n"
+ " {R10 += asl(R11,#2); R11 = and(%2,#0x1f)}\n"
+ "1: R12 = memw_locked(R10);\n"
+ " { P0 = tstbit(R12,R11); R12 = clrbit(R12,R11); }\n"
+ " memw_locked(R10,P1) = R12;\n"
+ " {if !P1 jump 1b; %0 = mux(P0,#1,#0);}\n"
+ : "=&r" (oldval)
+ : "r" (addr), "r" (nr)
+ : "r10", "r11", "r12", "p0", "p1", "memory"
+ );
+
+ return oldval;
+}
+
+/**
+ * test_and_set_bit - set a bit and return its old value
+ * @nr: bit number to set
+ * @addr: pointer to memory
+ */
+static inline int test_and_set_bit(int nr, volatile void *addr)
+{
+ int oldval;
+
+ __asm__ __volatile__ (
+ " {R10 = %1; R11 = asr(%2,#5); }\n"
+ " {R10 += asl(R11,#2); R11 = and(%2,#0x1f)}\n"
+ "1: R12 = memw_locked(R10);\n"
+ " { P0 = tstbit(R12,R11); R12 = setbit(R12,R11); }\n"
+ " memw_locked(R10,P1) = R12;\n"
+ " {if !P1 jump 1b; %0 = mux(P0,#1,#0);}\n"
+ : "=&r" (oldval)
+ : "r" (addr), "r" (nr)
+ : "r10", "r11", "r12", "p0", "p1", "memory"
+ );
+
+
+ return oldval;
+
+}
+
+/**
+ * test_and_change_bit - toggle a bit and return its old value
+ * @nr: bit number to set
+ * @addr: pointer to memory
+ */
+static inline int test_and_change_bit(int nr, volatile void *addr)
+{
+ int oldval;
+
+ __asm__ __volatile__ (
+ " {R10 = %1; R11 = asr(%2,#5); }\n"
+ " {R10 += asl(R11,#2); R11 = and(%2,#0x1f)}\n"
+ "1: R12 = memw_locked(R10);\n"
+ " { P0 = tstbit(R12,R11); R12 = togglebit(R12,R11); }\n"
+ " memw_locked(R10,P1) = R12;\n"
+ " {if !P1 jump 1b; %0 = mux(P0,#1,#0);}\n"
+ : "=&r" (oldval)
+ : "r" (addr), "r" (nr)
+ : "r10", "r11", "r12", "p0", "p1", "memory"
+ );
+
+ return oldval;
+
+}
+
+/*
+ * Atomic, but doesn't care about the return value.
+ * Rewrite later to save a cycle or two.
+ */
+
+static inline void clear_bit(int nr, volatile void *addr)
+{
+ test_and_clear_bit(nr, addr);
+}
+
+static inline void set_bit(int nr, volatile void *addr)
+{
+ test_and_set_bit(nr, addr);
+}
+
+static inline void change_bit(int nr, volatile void *addr)
+{
+ test_and_change_bit(nr, addr);
+}
+
+
+/*
+ * These are allowed to be non-atomic. In fact the generic flavors are
+ * in non-atomic.h. Would it be better to use intrinsics for this?
+ *
+ * OK, writes in our architecture do not invalidate LL/SC, so this has to
+ * be atomic, particularly for things like slab_lock and slab_unlock.
+ *
+ */
+static inline void __clear_bit(int nr, volatile unsigned long *addr)
+{
+ test_and_clear_bit(nr, addr);
+}
+
+static inline void __set_bit(int nr, volatile unsigned long *addr)
+{
+ test_and_set_bit(nr, addr);
+}
+
+static inline void __change_bit(int nr, volatile unsigned long *addr)
+{
+ test_and_change_bit(nr, addr);
+}
+
+/* Apparently, at least some of these are allowed to be non-atomic */
+static inline int __test_and_clear_bit(int nr, volatile unsigned long *addr)
+{
+ return test_and_clear_bit(nr, addr);
+}
+
+static inline int __test_and_set_bit(int nr, volatile unsigned long *addr)
+{
+ return test_and_set_bit(nr, addr);
+}
+
+static inline int __test_and_change_bit(int nr, volatile unsigned long *addr)
+{
+ return test_and_change_bit(nr, addr);
+}
+
+static inline int __test_bit(int nr, const volatile unsigned long *addr)
+{
+ int retval;
+
+ asm volatile(
+ "{P0 = tstbit(%1,%2); if (P0.new) %0 = #1; if (!P0.new) %0 = #0;}\n"
+ : "=&r" (retval)
+ : "r" (addr[BIT_WORD(nr)]), "r" (nr % BITS_PER_LONG)
+ : "p0"
+ );
+
+ return retval;
+}
+
+#define test_bit(nr, addr) __test_bit(nr, addr)
+
+/*
+ * ffz - find first zero in word.
+ * @word: The word to search
+ *
+ * Undefined if no zero exists, so code should check against ~0UL first.
+ */
+static inline long ffz(int x)
+{
+ int r;
+
+ asm("%0 = ct1(%1);\n"
+ : "=&r" (r)
+ : "r" (x));
+ return r;
+}
+
+/*
+ * fls - find last (most-significant) bit set
+ * @x: the word to search
+ *
+ * This is defined the same way as ffs.
+ * Note fls(0) = 0, fls(1) = 1, fls(0x80000000) = 32.
+ */
+static inline long fls(int x)
+{
+ int r;
+
+ asm("{ %0 = cl0(%1);}\n"
+ "%0 = sub(#32,%0);\n"
+ : "=&r" (r)
+ : "r" (x)
+ : "p0");
+
+ return r;
+}
+
+/*
+ * ffs - find first bit set
+ * @x: the word to search
+ *
+ * This is defined the same way as
+ * the libc and compiler builtin ffs routines, therefore
+ * differs in spirit from the above ffz (man ffs).
+ */
+static inline long ffs(int x)
+{
+ int r;
+
+ asm("{ P0 = cmp.eq(%1,#0); %0 = ct0(%1);}\n"
+ "{ if P0 %0 = #0; if !P0 %0 = add(%0,#1);}\n"
+ : "=&r" (r)
+ : "r" (x)
+ : "p0");
+
+ return r;
+}
+
+/*
+ * __ffs - find first bit in word.
+ * @word: The word to search
+ *
+ * Undefined if no bit exists, so code should check against 0 first.
+ *
+ * bits_per_long assumed to be 32
+ * numbering starts at 0 I think (instead of 1 like ffs)
+ */
+static inline unsigned long __ffs(unsigned long word)
+{
+ int num;
+
+ asm("%0 = ct0(%1);\n"
+ : "=&r" (num)
+ : "r" (word));
+
+ return num;
+}
+
+/*
+ * __fls - find last (most-significant) set bit in a long word
+ * @word: the word to search
+ *
+ * Undefined if no set bit exists, so code should check against 0 first.
+ * bits_per_long assumed to be 32
+ */
+static inline unsigned long __fls(unsigned long word)
+{
+ int num;
+
+ asm("%0 = cl0(%1);\n"
+ "%0 = sub(#31,%0);\n"
+ : "=&r" (num)
+ : "r" (word));
+
+ return num;
+}
+
+#include <asm-generic/bitops/lock.h>
+#include <asm-generic/bitops/find.h>
+
+#include <asm-generic/bitops/fls64.h>
+#include <asm-generic/bitops/sched.h>
+#include <asm-generic/bitops/hweight.h>
+
+#include <asm-generic/bitops/le.h>
+#include <asm-generic/bitops/ext2-atomic.h>
+
+#endif /* __KERNEL__ */
+#endif
--
1.7.1
--
Sent by an employee of the Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum.
^ permalink raw reply related
* [PATCH v5 02/36] Hexagon: Core arch-specific header files
From: Richard Kuo @ 2011-10-31 23:27 UTC (permalink / raw)
To: torvalds, linux-kernel
Signed-off-by: Richard Kuo <rkuo@codeaurora.org>
Signed-off-by: Linas Vepstas <linas@codeaurora.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
---
arch/hexagon/include/asm/byteorder.h | 28 ++++
arch/hexagon/include/asm/elf.h | 229 +++++++++++++++++++++++++++++++++
arch/hexagon/include/asm/fpu.h | 4 +
arch/hexagon/include/asm/futex.h | 137 ++++++++++++++++++++
arch/hexagon/include/asm/linkage.h | 25 ++++
arch/hexagon/include/asm/module.h | 26 ++++
arch/hexagon/include/asm/param.h | 26 ++++
arch/hexagon/include/asm/perf_event.h | 24 ++++
arch/hexagon/include/asm/string.h | 32 +++++
arch/hexagon/include/asm/suspend.h | 27 ++++
arch/hexagon/include/asm/swab.h | 24 ++++
arch/hexagon/include/asm/timex.h | 37 ++++++
12 files changed, 619 insertions(+), 0 deletions(-)
create mode 100644 arch/hexagon/include/asm/byteorder.h
create mode 100644 arch/hexagon/include/asm/elf.h
create mode 100644 arch/hexagon/include/asm/fpu.h
create mode 100644 arch/hexagon/include/asm/futex.h
create mode 100644 arch/hexagon/include/asm/linkage.h
create mode 100644 arch/hexagon/include/asm/module.h
create mode 100644 arch/hexagon/include/asm/param.h
create mode 100644 arch/hexagon/include/asm/perf_event.h
create mode 100644 arch/hexagon/include/asm/string.h
create mode 100644 arch/hexagon/include/asm/suspend.h
create mode 100644 arch/hexagon/include/asm/swab.h
create mode 100644 arch/hexagon/include/asm/timex.h
diff --git a/arch/hexagon/include/asm/byteorder.h b/arch/hexagon/include/asm/byteorder.h
new file mode 100644
index 0000000..0e19b9f
--- /dev/null
+++ b/arch/hexagon/include/asm/byteorder.h
@@ -0,0 +1,28 @@
+/*
+ * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+#ifndef _ASM_BYTEORDER_H
+#define _ASM_BYTEORDER_H
+
+#if defined(__GNUC__) && !defined(__STRICT_ANSI__) || defined(__KERNEL__)
+# define __BYTEORDER_HAS_U64__
+#endif
+
+#include <linux/byteorder/little_endian.h>
+
+#endif /* _ASM_BYTEORDER_H */
diff --git a/arch/hexagon/include/asm/elf.h b/arch/hexagon/include/asm/elf.h
new file mode 100644
index 0000000..37976a0
--- /dev/null
+++ b/arch/hexagon/include/asm/elf.h
@@ -0,0 +1,229 @@
+/*
+ * ELF definitions for the Hexagon architecture
+ *
+ * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+#ifndef __ASM_ELF_H
+#define __ASM_ELF_H
+
+#include <asm/ptrace.h>
+#include <asm/user.h>
+
+/*
+ * This should really be in linux/elf-em.h.
+ */
+#define EM_HEXAGON 164 /* QUALCOMM Hexagon */
+
+struct elf32_hdr;
+
+/*
+ * ELF header e_flags defines.
+ */
+
+/* should have stuff like "CPU type" and maybe "ABI version", etc */
+
+/* Hexagon relocations */
+ /* V2 */
+#define R_HEXAGON_NONE 0
+#define R_HEXAGON_B22_PCREL 1
+#define R_HEXAGON_B15_PCREL 2
+#define R_HEXAGON_B7_PCREL 3
+#define R_HEXAGON_LO16 4
+#define R_HEXAGON_HI16 5
+#define R_HEXAGON_32 6
+#define R_HEXAGON_16 7
+#define R_HEXAGON_8 8
+#define R_HEXAGON_GPREL16_0 9
+#define R_HEXAGON_GPREL16_1 10
+#define R_HEXAGON_GPREL16_2 11
+#define R_HEXAGON_GPREL16_3 12
+#define R_HEXAGON_HL16 13
+ /* V3 */
+#define R_HEXAGON_B13_PCREL 14
+ /* V4 */
+#define R_HEXAGON_B9_PCREL 15
+ /* V4 (extenders) */
+#define R_HEXAGON_B32_PCREL_X 16
+#define R_HEXAGON_32_6_X 17
+ /* V4 (extended) */
+#define R_HEXAGON_B22_PCREL_X 18
+#define R_HEXAGON_B15_PCREL_X 19
+#define R_HEXAGON_B13_PCREL_X 20
+#define R_HEXAGON_B9_PCREL_X 21
+#define R_HEXAGON_B7_PCREL_X 22
+#define R_HEXAGON_16_X 23
+#define R_HEXAGON_12_X 24
+#define R_HEXAGON_11_X 25
+#define R_HEXAGON_10_X 26
+#define R_HEXAGON_9_X 27
+#define R_HEXAGON_8_X 28
+#define R_HEXAGON_7_X 29
+#define R_HEXAGON_6_X 30
+ /* V2 PIC */
+#define R_HEXAGON_32_PCREL 31
+#define R_HEXAGON_COPY 32
+#define R_HEXAGON_GLOB_DAT 33
+#define R_HEXAGON_JMP_SLOT 34
+#define R_HEXAGON_RELATIVE 35
+#define R_HEXAGON_PLT_B22_PCREL 36
+#define R_HEXAGON_GOTOFF_LO16 37
+#define R_HEXAGON_GOTOFF_HI16 38
+#define R_HEXAGON_GOTOFF_32 39
+#define R_HEXAGON_GOT_LO16 40
+#define R_HEXAGON_GOT_HI16 41
+#define R_HEXAGON_GOT_32 42
+#define R_HEXAGON_GOT_16 43
+
+/*
+ * ELF register definitions..
+ */
+typedef unsigned long elf_greg_t;
+
+typedef struct user_regs_struct elf_gregset_t;
+#define ELF_NGREG (sizeof(elf_gregset_t)/sizeof(unsigned long))
+
+/* Placeholder */
+typedef unsigned long elf_fpregset_t;
+
+/*
+ * Bypass the whole "regsets" thing for now and use the define.
+ */
+
+#define ELF_CORE_COPY_REGS(DEST, REGS) \
+do { \
+ DEST.r0 = REGS->r00; \
+ DEST.r1 = REGS->r01; \
+ DEST.r2 = REGS->r02; \
+ DEST.r3 = REGS->r03; \
+ DEST.r4 = REGS->r04; \
+ DEST.r5 = REGS->r05; \
+ DEST.r6 = REGS->r06; \
+ DEST.r7 = REGS->r07; \
+ DEST.r8 = REGS->r08; \
+ DEST.r9 = REGS->r09; \
+ DEST.r10 = REGS->r10; \
+ DEST.r11 = REGS->r11; \
+ DEST.r12 = REGS->r12; \
+ DEST.r13 = REGS->r13; \
+ DEST.r14 = REGS->r14; \
+ DEST.r15 = REGS->r15; \
+ DEST.r16 = REGS->r16; \
+ DEST.r17 = REGS->r17; \
+ DEST.r18 = REGS->r18; \
+ DEST.r19 = REGS->r19; \
+ DEST.r20 = REGS->r20; \
+ DEST.r21 = REGS->r21; \
+ DEST.r22 = REGS->r22; \
+ DEST.r23 = REGS->r23; \
+ DEST.r24 = REGS->r24; \
+ DEST.r25 = REGS->r25; \
+ DEST.r26 = REGS->r26; \
+ DEST.r27 = REGS->r27; \
+ DEST.r28 = REGS->r28; \
+ DEST.r29 = pt_psp(REGS); \
+ DEST.r30 = REGS->r30; \
+ DEST.r31 = REGS->r31; \
+ DEST.sa0 = REGS->sa0; \
+ DEST.lc0 = REGS->lc0; \
+ DEST.sa1 = REGS->sa1; \
+ DEST.lc1 = REGS->lc1; \
+ DEST.m0 = REGS->m0; \
+ DEST.m1 = REGS->m1; \
+ DEST.usr = REGS->usr; \
+ DEST.p3_0 = REGS->preds; \
+ DEST.gp = REGS->gp; \
+ DEST.ugp = REGS->ugp; \
+ DEST.pc = pt_elr(REGS); \
+ DEST.cause = pt_cause(REGS); \
+ DEST.badva = pt_badva(REGS); \
+} while (0);
+
+
+
+/*
+ * This is used to ensure we don't load something for the wrong architecture.
+ * Checks the machine and ABI type.
+ */
+#define elf_check_arch(hdr) ((hdr)->e_machine == EM_HEXAGON)
+
+/*
+ * These are used to set parameters in the core dumps.
+ */
+#define ELF_CLASS ELFCLASS32
+#define ELF_DATA ELFDATA2LSB
+#define ELF_ARCH EM_HEXAGON
+
+#ifdef CONFIG_HEXAGON_ARCH_V2
+#define ELF_CORE_EFLAGS 0x1
+#endif
+
+#ifdef CONFIG_HEXAGON_ARCH_V3
+#define ELF_CORE_EFLAGS 0x2
+#endif
+
+#ifdef CONFIG_HEXAGON_ARCH_V4
+#define ELF_CORE_EFLAGS 0x3
+#endif
+
+/*
+ * Some architectures have ld.so set up a pointer to a function
+ * to be registered using atexit, to facilitate cleanup. So that
+ * static executables will be well-behaved, we would null the register
+ * in question here, in the pt_regs structure passed. For now,
+ * leave it a null macro.
+ */
+#define ELF_PLAT_INIT(regs, load_addr) do { } while (0)
+
+#define USE_ELF_CORE_DUMP
+#define CORE_DUMP_USE_REGSET
+
+/* Hrm is this going to cause problems for changing PAGE_SIZE? */
+#define ELF_EXEC_PAGESIZE 4096
+
+/*
+ * This is the location that an ET_DYN program is loaded if exec'ed. Typical
+ * use of this is to invoke "./ld.so someprog" to test out a new version of
+ * the loader. We need to make sure that it is out of the way of the program
+ * that it will "exec", and that there is sufficient room for the brk.
+ */
+#define ELF_ET_DYN_BASE 0x08000000UL
+
+/*
+ * This yields a mask that user programs can use to figure out what
+ * instruction set this cpu supports.
+ */
+#define ELF_HWCAP (0)
+
+/*
+ * This yields a string that ld.so will use to load implementation
+ * specific libraries for optimization. This is more specific in
+ * intent than poking at uname or /proc/cpuinfo.
+ */
+#define ELF_PLATFORM (NULL)
+
+#ifdef __KERNEL__
+#define SET_PERSONALITY(ex) set_personality(PER_LINUX)
+#endif
+
+#define ARCH_HAS_SETUP_ADDITIONAL_PAGES 1
+struct linux_binprm;
+extern int arch_setup_additional_pages(struct linux_binprm *bprm,
+ int uses_interp);
+
+
+#endif
diff --git a/arch/hexagon/include/asm/fpu.h b/arch/hexagon/include/asm/fpu.h
new file mode 100644
index 0000000..0e135ea
--- /dev/null
+++ b/arch/hexagon/include/asm/fpu.h
@@ -0,0 +1,4 @@
+/*
+ * If the FPU is used inside the kernel,
+ * kernel_fpu_end() will be defined here.
+ */
diff --git a/arch/hexagon/include/asm/futex.h b/arch/hexagon/include/asm/futex.h
new file mode 100644
index 0000000..7e597f8
--- /dev/null
+++ b/arch/hexagon/include/asm/futex.h
@@ -0,0 +1,137 @@
+#ifndef _ASM_HEXAGON_FUTEX_H
+#define _ASM_HEXAGON_FUTEX_H
+
+#ifdef __KERNEL__
+
+#include <linux/futex.h>
+#include <linux/uaccess.h>
+#include <asm/errno.h>
+
+/* XXX TODO-- need to add sync barriers! */
+
+#define __futex_atomic_op(insn, ret, oldval, uaddr, oparg) \
+ __asm__ __volatile( \
+ "1: %0 = memw_locked(%3);\n" \
+ /* For example: %1 = %4 */ \
+ insn \
+ "2: memw_locked(%3,p2) = %1;\n" \
+ " if !p2 jump 1b;\n" \
+ " %1 = #0;\n" \
+ "3:\n" \
+ ".section .fixup,\"ax\"\n" \
+ "4: %1 = #%5;\n" \
+ " jump 3b\n" \
+ ".previous\n" \
+ ".section __ex_table,\"a\"\n" \
+ ".long 1b,4b,2b,4b\n" \
+ ".previous\n" \
+ : "=&r" (oldval), "=&r" (ret), "+m" (*uaddr) \
+ : "r" (uaddr), "r" (oparg), "i" (-EFAULT) \
+ : "p2", "memory")
+
+
+static inline int
+futex_atomic_op_inuser(int encoded_op, int __user *uaddr)
+{
+ int op = (encoded_op >> 28) & 7;
+ int cmp = (encoded_op >> 24) & 15;
+ int oparg = (encoded_op << 8) >> 20;
+ int cmparg = (encoded_op << 20) >> 20;
+ int oldval = 0, ret;
+ if (encoded_op & (FUTEX_OP_OPARG_SHIFT << 28))
+ oparg = 1 << oparg;
+
+ if (!access_ok(VERIFY_WRITE, uaddr, sizeof(int)))
+ return -EFAULT;
+
+ pagefault_disable();
+
+ switch (op) {
+ case FUTEX_OP_SET:
+ __futex_atomic_op("%1 = %4\n", ret, oldval, uaddr, oparg);
+ break;
+ case FUTEX_OP_ADD:
+ __futex_atomic_op("%1 = add(%0,%4)\n", ret, oldval, uaddr,
+ oparg);
+ break;
+ case FUTEX_OP_OR:
+ __futex_atomic_op("%1 = or(%0,%4)\n", ret, oldval, uaddr,
+ oparg);
+ break;
+ case FUTEX_OP_ANDN:
+ __futex_atomic_op("%1 = not(%4); %1 = and(%0,%1)\n", ret,
+ oldval, uaddr, oparg);
+ break;
+ case FUTEX_OP_XOR:
+ __futex_atomic_op("%1 = xor(%0,%4)\n", ret, oldval, uaddr,
+ oparg);
+ break;
+ default:
+ ret = -ENOSYS;
+ }
+
+ pagefault_enable();
+
+ if (!ret) {
+ switch (cmp) {
+ case FUTEX_OP_CMP_EQ:
+ ret = (oldval == cmparg);
+ break;
+ case FUTEX_OP_CMP_NE:
+ ret = (oldval != cmparg);
+ break;
+ case FUTEX_OP_CMP_LT:
+ ret = (oldval < cmparg);
+ break;
+ case FUTEX_OP_CMP_GE:
+ ret = (oldval >= cmparg);
+ break;
+ case FUTEX_OP_CMP_LE:
+ ret = (oldval <= cmparg);
+ break;
+ case FUTEX_OP_CMP_GT:
+ ret = (oldval > cmparg);
+ break;
+ default:
+ ret = -ENOSYS;
+ }
+ }
+ return ret;
+}
+
+static inline int
+futex_atomic_cmpxchg_inatomic(u32 *uval, u32 __user *uaddr, u32 oldval,
+ u32 newval)
+{
+ int prev;
+ int ret;
+
+ if (!access_ok(VERIFY_WRITE, uaddr, sizeof(u32)))
+ return -EFAULT;
+
+ __asm__ __volatile__ (
+ "1: %1 = memw_locked(%3)\n"
+ " {\n"
+ " p2 = cmp.eq(%1,%4)\n"
+ " if !p2.new jump:NT 3f\n"
+ " }\n"
+ "2: memw_locked(%3,p2) = %5\n"
+ " if !p2 jump 1b\n"
+ "3:\n"
+ ".section .fixup,\"ax\"\n"
+ "4: %0 = #%6\n"
+ " jump 3b\n"
+ ".previous\n"
+ ".section __ex_table,\"a\"\n"
+ ".long 1b,4b,2b,4b\n"
+ ".previous\n"
+ : "+r" (ret), "=&r" (prev), "+m" (*uaddr)
+ : "r" (uaddr), "r" (oldval), "r" (newval), "i"(-EFAULT)
+ : "p2", "memory");
+
+ *uval = prev;
+ return ret;
+}
+
+#endif /* __KERNEL__ */
+#endif /* _ASM_HEXAGON_FUTEX_H */
diff --git a/arch/hexagon/include/asm/linkage.h b/arch/hexagon/include/asm/linkage.h
new file mode 100644
index 0000000..a00b85f
--- /dev/null
+++ b/arch/hexagon/include/asm/linkage.h
@@ -0,0 +1,25 @@
+/*
+ * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+#ifndef __ASM_LINKAGE_H
+#define __ASM_LINKAGE_H
+
+#define __ALIGN .align 4
+#define __ALIGN_STR ".align 4"
+
+#endif
diff --git a/arch/hexagon/include/asm/module.h b/arch/hexagon/include/asm/module.h
new file mode 100644
index 0000000..72ba494
--- /dev/null
+++ b/arch/hexagon/include/asm/module.h
@@ -0,0 +1,26 @@
+/*
+ * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+#ifndef _ASM_MODULE_H
+#define _ASM_MODULE_H
+
+#include <asm-generic/module.h>
+
+#define MODULE_ARCH_VERMAGIC __stringify(PROCESSOR_MODEL_NAME) " "
+
+#endif
diff --git a/arch/hexagon/include/asm/param.h b/arch/hexagon/include/asm/param.h
new file mode 100644
index 0000000..285344b
--- /dev/null
+++ b/arch/hexagon/include/asm/param.h
@@ -0,0 +1,26 @@
+/*
+ * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+#ifndef _ASM_PARAM_H
+#define _ASM_PARAM_H
+
+#define EXEC_PAGESIZE 16384
+
+#include <asm-generic/param.h>
+
+#endif
diff --git a/arch/hexagon/include/asm/perf_event.h b/arch/hexagon/include/asm/perf_event.h
new file mode 100644
index 0000000..6c2910f
--- /dev/null
+++ b/arch/hexagon/include/asm/perf_event.h
@@ -0,0 +1,24 @@
+/*
+ * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+#ifndef _ASM_PERF_EVENT_H
+#define _ASM_PERF_EVENT_H
+
+#define PERF_EVENT_INDEX_OFFSET 0
+
+#endif /* _ASM_PERF_EVENT_H */
diff --git a/arch/hexagon/include/asm/string.h b/arch/hexagon/include/asm/string.h
new file mode 100644
index 0000000..f4489c1
--- /dev/null
+++ b/arch/hexagon/include/asm/string.h
@@ -0,0 +1,32 @@
+/*
+ * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+#ifndef _ASM_STRING_H_
+#define _ASM_STRING_H_
+
+#ifdef __KERNEL__
+#define __HAVE_ARCH_MEMCPY
+extern void *memcpy(void *__to, __const__ void *__from, size_t __n);
+
+/* ToDo: use dczeroa, accelerate the compiler-constant zero case */
+#define __HAVE_ARCH_MEMSET
+extern void *memset(void *__to, int c, size_t __n);
+#endif
+
+
+#endif /* _ASM_STRING_H_ */
diff --git a/arch/hexagon/include/asm/suspend.h b/arch/hexagon/include/asm/suspend.h
new file mode 100644
index 0000000..089dd82
--- /dev/null
+++ b/arch/hexagon/include/asm/suspend.h
@@ -0,0 +1,27 @@
+/*
+ * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+#ifndef _ASM_SUSPEND_H
+#define _ASM_SUSPEND_H
+
+static inline int arch_prepare_suspend(void)
+{
+ return 0;
+}
+
+#endif
diff --git a/arch/hexagon/include/asm/swab.h b/arch/hexagon/include/asm/swab.h
new file mode 100644
index 0000000..99cf0be
--- /dev/null
+++ b/arch/hexagon/include/asm/swab.h
@@ -0,0 +1,24 @@
+/*
+ * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+#ifndef _ASM_SWAB_H
+#define _ASM_SWAB_H
+
+#define __SWAB_64_THRU_32__
+
+#endif
diff --git a/arch/hexagon/include/asm/timex.h b/arch/hexagon/include/asm/timex.h
new file mode 100644
index 0000000..09cd843
--- /dev/null
+++ b/arch/hexagon/include/asm/timex.h
@@ -0,0 +1,37 @@
+/*
+ * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+#ifndef _ASM_TIMEX_H
+#define _ASM_TIMEX_H
+
+#include <asm-generic/timex.h>
+#include <asm/timer-regs.h>
+
+/* Using TCX0 as our clock. CLOCK_TICK_RATE scheduled to be removed. */
+#define CLOCK_TICK_RATE TCX0_CLK_RATE
+
+#define ARCH_HAS_READ_CURRENT_TIMER
+
+static inline int read_current_timer(unsigned long *timer_val)
+{
+ *timer_val = (unsigned long) __vmgettime();
+ return 0;
+}
+
+#endif
+
--
1.7.1
--
Sent by an employee of the Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum.
^ permalink raw reply related
* [PATCH v5 01/36] Hexagon: Add generic headers
From: Richard Kuo @ 2011-10-31 23:26 UTC (permalink / raw)
To: torvalds, linux-kernel
Signed-off-by: Richard Kuo <rkuo@codeaurora.org>
Signed-off-by: Linas Vepstas <linas@codeaurora.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
---
arch/hexagon/include/asm/Kbuild | 59 +++++++++++++++++++++++
arch/hexagon/include/asm/bitsperlong.h | 26 ++++++++++
arch/hexagon/include/asm/mutex.h | 8 +++
arch/hexagon/include/asm/setup.h | 29 +++++++++++
arch/hexagon/include/asm/user.h | 81 ++++++++++++++++++++++++++++++++
5 files changed, 203 insertions(+), 0 deletions(-)
create mode 100644 arch/hexagon/include/asm/Kbuild
create mode 100644 arch/hexagon/include/asm/bitsperlong.h
create mode 100644 arch/hexagon/include/asm/mutex.h
create mode 100644 arch/hexagon/include/asm/setup.h
create mode 100644 arch/hexagon/include/asm/user.h
diff --git a/arch/hexagon/include/asm/Kbuild b/arch/hexagon/include/asm/Kbuild
new file mode 100644
index 0000000..6219617
--- /dev/null
+++ b/arch/hexagon/include/asm/Kbuild
@@ -0,0 +1,59 @@
+include include/asm-generic/Kbuild.asm
+
+header-y += registers.h
+header-y += ucontext.h
+header-y += user.h
+
+generic-y += auxvec.h
+generic-y += bug.h
+generic-y += bugs.h
+generic-y += cpumask.h
+generic-y += cputime.h
+generic-y += current.h
+generic-y += device.h
+generic-y += div64.h
+generic-y += emergency-restart.h
+generic-y += errno.h
+generic-y += fb.h
+generic-y += fcntl.h
+generic-y += ftrace.h
+generic-y += hardirq.h
+generic-y += hw_irq.h
+generic-y += ioctl.h
+generic-y += ioctls.h
+generic-y += iomap.h
+generic-y += ipcbuf.h
+generic-y += ipc.h
+generic-y += irq_regs.h
+generic-y += kdebug.h
+generic-y += kmap_types.h
+generic-y += local64.h
+generic-y += local.h
+generic-y += local.h
+generic-y += mman.h
+generic-y += msgbuf.h
+generic-y += pci.h
+generic-y += percpu.h
+generic-y += poll.h
+generic-y += posix_types.h
+generic-y += resource.h
+generic-y += rwsem.h
+generic-y += scatterlist.h
+generic-y += sections.h
+generic-y += segment.h
+generic-y += sembuf.h
+generic-y += shmbuf.h
+generic-y += shmparam.h
+generic-y += siginfo.h
+generic-y += socket.h
+generic-y += sockios.h
+generic-y += statfs.h
+generic-y += stat.h
+generic-y += termbits.h
+generic-y += termios.h
+generic-y += topology.h
+generic-y += types.h
+generic-y += ucontext.h
+generic-y += unaligned.h
+generic-y += xor.h
+
diff --git a/arch/hexagon/include/asm/bitsperlong.h b/arch/hexagon/include/asm/bitsperlong.h
new file mode 100644
index 0000000..2701cae
--- /dev/null
+++ b/arch/hexagon/include/asm/bitsperlong.h
@@ -0,0 +1,26 @@
+/*
+ * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+#ifndef __ASM_HEXAGON_BITSPERLONG_H
+#define __ASM_HEXAGON_BITSPERLONG_H
+
+#define __BITS_PER_LONG 32
+
+#include <asm-generic/bitsperlong.h>
+
+#endif
diff --git a/arch/hexagon/include/asm/mutex.h b/arch/hexagon/include/asm/mutex.h
new file mode 100644
index 0000000..58b52de
--- /dev/null
+++ b/arch/hexagon/include/asm/mutex.h
@@ -0,0 +1,8 @@
+/*
+ * Pull in the generic implementation for the mutex fastpath.
+ *
+ * TODO: implement optimized primitives instead, or leave the generic
+ * implementation in place, or pick the atomic_xchg() based generic
+ * implementation. (see asm-generic/mutex-xchg.h for details)
+ */
+#include <asm-generic/mutex-xchg.h>
diff --git a/arch/hexagon/include/asm/setup.h b/arch/hexagon/include/asm/setup.h
new file mode 100644
index 0000000..3b754c5
--- /dev/null
+++ b/arch/hexagon/include/asm/setup.h
@@ -0,0 +1,29 @@
+/*
+ * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+#ifndef _ASM_SETUP_H
+#define _ASM_SETUP_H
+
+#include <linux/init.h>
+#include <asm-generic/setup.h>
+
+extern char external_cmdline_buffer;
+
+void __init setup_arch_memory(void);
+
+#endif
diff --git a/arch/hexagon/include/asm/user.h b/arch/hexagon/include/asm/user.h
new file mode 100644
index 0000000..3a55078
--- /dev/null
+++ b/arch/hexagon/include/asm/user.h
@@ -0,0 +1,81 @@
+/*
+ * Copyright (c) 2010-2011, Code Aurora Forum. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA.
+ */
+
+#ifndef HEXAGON_ASM_USER_H
+#define HEXAGON_ASM_USER_H
+
+/*
+ * Layout for registers passed in elf core dumps to userspace.
+ *
+ * Basically a rearranged subset of "pt_regs".
+ *
+ * Interested parties: libc, gdb...
+ */
+
+struct user_regs_struct {
+ unsigned long r0;
+ unsigned long r1;
+ unsigned long r2;
+ unsigned long r3;
+ unsigned long r4;
+ unsigned long r5;
+ unsigned long r6;
+ unsigned long r7;
+ unsigned long r8;
+ unsigned long r9;
+ unsigned long r10;
+ unsigned long r11;
+ unsigned long r12;
+ unsigned long r13;
+ unsigned long r14;
+ unsigned long r15;
+ unsigned long r16;
+ unsigned long r17;
+ unsigned long r18;
+ unsigned long r19;
+ unsigned long r20;
+ unsigned long r21;
+ unsigned long r22;
+ unsigned long r23;
+ unsigned long r24;
+ unsigned long r25;
+ unsigned long r26;
+ unsigned long r27;
+ unsigned long r28;
+ unsigned long r29;
+ unsigned long r30;
+ unsigned long r31;
+ unsigned long sa0;
+ unsigned long lc0;
+ unsigned long sa1;
+ unsigned long lc1;
+ unsigned long m0;
+ unsigned long m1;
+ unsigned long usr;
+ unsigned long p3_0;
+ unsigned long gp;
+ unsigned long ugp;
+ unsigned long pc;
+ unsigned long cause;
+ unsigned long badva;
+ unsigned long pad1; /* pad out to 48 words total */
+ unsigned long pad2; /* pad out to 48 words total */
+ unsigned long pad3; /* pad out to 48 words total */
+};
+
+#endif
--
1.7.1
--
Sent by an employee of the Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum.
^ permalink raw reply related
* Re: [msysGit] Re: [PATCHv2] Compile fix for MSVC
From: Johannes Schindelin @ 2011-11-01 0:09 UTC (permalink / raw)
To: Junio C Hamano; +Cc: Vincent van Ravesteijn, git, kusmabite, ramsay, msysgit
In-Reply-To: <7vd3dcq4s5.fsf@alter.siamese.dyndns.org>
Hi,
On Mon, 31 Oct 2011, Junio C Hamano wrote:
> Thanks. The patch looks good from a POSIX person's point of view, and I
> do not immediately see how this would break other variants of Windows
> build at least from the code inspection.
As my virtual machine still ran the test suite after my latest rebasing
merge when I left work, I could not test the MSVC stuff just yet. I wanted
to do that tomorrow and either merge or come back with suggestions.
Ciao,
Johannes
^ permalink raw reply
* Re: [Qemu-devel] GSoC mentor summit QEMU users session
From: Alexander Graf @ 2011-11-01 0:08 UTC (permalink / raw)
To: Peter Maydell; +Cc: qemu-devel@nongnu.org Developers
In-Reply-To: <CAFEAcA9jh5E5WQH6YwAX1HBKg8W5NB5aR8kmTTz2h6wdRhc0KA@mail.gmail.com>
On 31.10.2011, at 06:12, Peter Maydell <peter.maydell@linaro.org> wrote:
> On 29 October 2011 14:52, Alexander Graf <agraf@suse.de> wrote:
>> We should also show people unmaintained areas. The conclusion was a wiki
>> page with subsystems and status so people know what to expect. Maybe we
>> could generate this from the MAINTAINERS file?
>
> Sounds like a good idea, although I think we might need to expand
> MAINTAINERS a bit -- I get the impression that there are a lot of
> "little bits" that fall into the gaps between the top-level areas
> marked out in MAINTAINERS.
True. We do however have file path matches, so we could easily find unmaintained files.
>
>> Also, an easy way of counterfighting the feeling ignored part is
>> to tell people that they just hit an unmaintained area. There's
>> nothing more frustrating than sending a patch and get no reply.
>> Receiving a reply "Sorry, this area is unmaintained. Please find
>> someone to review it." would already be enough for most people.
>
> The difficulty that strikes me with this is that I'm not sure any
> one person can reliably look at a patch and say "that's for an
> unmaintained area" (at least, I know what areas I can review but
> I have no idea about everybody else...) So you can only really
> tell by default, ie if the patch sits for a few weeks without
> any reply...
See above. I think we could script this :)
>
>> A lot of people seem to also have code that doesn't make sense
>> upstream, for example implementing a one-off device that only
>> really matters for their own devboard which nobody else owns.
>> For such cases, having a plugin framework would be handy. I
>> interestingly enough got into the same discussion on LinuxCon
>> with some QEMU downstreams.
>
> If we get the qdev rework done then I think we're probably in
> a better position to have a plugin framework for devices. (There
> are some issues about API and ABI stability guarantees, of course.)
I'm not sure why we should. We could just follow the Linux kernel model and merely expose what's there. New version means new API.
Remember, I don't want this for commercial fire-and-forget device models. I want it for stuff that's either too unclean or too useless for upstream :).
Alex
>
^ permalink raw reply
* Agenda: Yocto Technical Team Meeting - Tuesday, November 01, 2011 8:00 AM-9:00 AM (UTC-08:00) Pacific Time (US & Canada)
From: Liu, Song @ 2011-11-01 0:07 UTC (permalink / raw)
To: Liu, Song, 'McCombe, Kevin', Wold, Saul, Zhang, Jessica,
'deVries, Alex', 'Polk, Jeffrey',
'Hatle, Mark', Purdie, Richard, Li, Susie, Lock, Joshua,
'Bruce Ashfield', Zanussi, Tom, Hart, Darren,
Flanagan, Elizabeth, Stewart, David C,
'Lieu.Ta@windriver.com',
'paul.anderson@windriver.com', Osier-mixon, Jeffrey,
Dmytriyenko, Denys, Kooi, Koen,
'yocto-ab@yoctoproject.org', Erway, Tracey M,
'Daniel Cauchy', Wang, Shane, 'Kridner, Jason',
Yu, Ke, Rifenbark, Scott M, 'Dixon, Brad',
'Jeremy Puhlman', yocto@yoctoproject.org,
'shsift-yocto@yahoo.com'
Cc: 'Ruff, Nithya', 'morvek@mvista.com'
In-Reply-To: <7E680CB2A76BB8438D08CF13C09FC53A16EBA481@FMSMSX151.amr.corp.intel.com>
* Opens collection - 5 min (Song)
* Yocto 1.0.2 and 1.1.1 point release update - 5 min (Josh)
* Yocto 1.2 planning and M1 planning - 20 min (team)
https://wiki.yoctoproject.org/wiki/Yocto_1.2_Schedule
https://wiki.yoctoproject.org/wiki/Yocto_1.2_Features
* Opens - 10 min
* Weekly team sharing (20 min)
-----Original Appointment-----
From: Liu, Song
Sent: Tuesday, August 16, 2011 3:27 PM
To: 'McCombe, Kevin'; Wold, Saul; Zhang, Jessica; 'deVries, Alex'; 'Polk, Jeffrey'; 'Hatle, Mark'; Purdie, Richard; Li, Susie; Lock, Joshua; 'Bruce Ashfield'; Zanussi, Tom; Hart, Darren; Flanagan, Elizabeth; Stewart, David C; 'Lieu.Ta@windriver.com'; 'paul.anderson@windriver.com'; Osier-mixon, Jeffrey; Dmytriyenko, Denys; Kooi, Koen; 'yocto-ab@yoctoproject.org'; Erway, Tracey M; 'Daniel Cauchy'; Wang, Shane; 'Kridner, Jason'; Yu, Ke; Rifenbark, Scott M; 'Dixon, Brad'; 'Jeremy Puhlman'; yocto@yoctoproject.org; 'shsift-yocto@yahoo.com'
Cc: 'morvek@mvista.com'; 'Ruff, Nithya'; Moeller, Thorsten; Saxena, Rahul; 'Ranslam, Rob'; 'Anders Darander'; Bodke, Kishore K
Subject: Yocto Technical Team Meeting
When: Tuesday, November 01, 2011 8:00 AM-9:00 AM (UTC-08:00) Pacific Time (US & Canada).
Where: Bridge Info Enclosed
Conference details
Conference name:
Yocto Technical Team
Conference date/start time:
Tue Jun 28, 2011 at 10:00 AM Central Daylight Time
Participants:
30
Duration:
60 minutes
Participant passcode:
49611427
Dial-in number:
1.972.995.7777
US Toll Free number:
1.877.561.6828
BlackBerry users, click this link to join your conference as a chairperson:
1.972.995.7777x67184230#
BlackBerry users, click this link to join your conference as a participant:
1.972.995.7777x49611427#
Depending on where you are dialing from, either your BlackBerry will pause and enter the passcode automatically or you will be prompted to click again to dial the passcode.
Local and Global Access Numbers
Country
Dial-in number
Australia:
1800 636 843
Czech Republic:
242 430 350
China (Beijing):
From office dial 8-9957777 or 8784277
Beijing Out of Office dial 5878 4277
China (Shanghai):
From office dial 8-9957777 or 3073322
Shanghai Out of Office dial 2307 3322
China (Shenzen):
From office dial 8-9957777 or 6007877
Shenzen Out of Office dial 2600 7877
China (Other Cities):
From IP phone dial 8-9957777
Other cities - Non IP phone dial 021-23073322
Denmark:
8060 1400
Finland:
09 41333477
France:
0497 275888
Germany:
08161 803232
Holland:
030 2417490
India:
BSNL subscribers use 1800 425 9996 (Toll Free)
Airtel subscribers use 0008 009 861 212 (Toll Free)
From TI Campus use 89957777
Others use 2509 9555 (Landline within Bangalore) or
80 2509 9555 (Outside Bangalore)
Israel:
09 790 6715
Italy:
039 69061234 (039 is local city code not country code)
Japan:
From TI Campus use 8 995 7777
Outside TI use 03 4331 3777
Malaysia:
From IP phone dial 2643799
From Kuala Lumpur dial 4264 3799
Outside Kuala Lumpur dial (03)4264 3799
Norway:
2 295 8744
Philippines:
From Baguio City use 4471177
From Metro Manila area use 8702477
Singapore:
From IP phone dial 3894777
Outside TI use 6389 4777
South Korea:
From IP phone dial 5606998
From Seoul dial 5606998
Outside Seoul dial (02)5606998
Sweden:
08 58755577
Taiwan:
From IP phone dial 1363
From Taipei dial 2241 1363
Outside Taipei dial (02)2241 1363
Turkey:
Landline Only dial 0811 288 0001
then enter 877 633 1123
UK:
01604 663003
US:
972 995 7777 or 1877 561 6828
Recurring conferences
First scheduled conference:
Tue Jun 28, 2011
Recurrence frequency:
Weekly - Every 1 week(s) on Tuesday
Recurrence ends:
End on Fri Jun 22, 2012, 10:40 AM CDT
^ permalink raw reply
* [patch 005/127] Cross Memory Attach
From: akpm @ 2011-11-01 0:06 UTC (permalink / raw)
To: torvalds
Cc: akpm, cyeoh, arnd, benh, dhowells, hpa, jmorris, linux-arch,
linux-man, mingo, paulus, tglx, yeohc
From: Christopher Yeoh <cyeoh@au1.ibm.com>
Subject: Cross Memory Attach
The basic idea behind cross memory attach is to allow MPI programs doing
intra-node communication to do a single copy of the message rather than a
double copy of the message via shared memory.
The following patch attempts to achieve this by allowing a destination
process, given an address and size from a source process, to copy memory
directly from the source process into its own address space via a system
call. There is also a symmetrical ability to copy from the current
process's address space into a destination process's address space.
- Use of /proc/pid/mem has been considered, but there are issues with
using it:
- Does not allow for specifying iovecs for both src and dest, assuming
preadv or pwritev was implemented either the area read from or
written to would need to be contiguous.
- Currently mem_read allows only processes who are currently
ptrace'ing the target and are still able to ptrace the target to read
from the target. This check could possibly be moved to the open call,
but its not clear exactly what race this restriction is stopping
(reason appears to have been lost)
- Having to send the fd of /proc/self/mem via SCM_RIGHTS on unix
domain socket is a bit ugly from a userspace point of view,
especially when you may have hundreds if not (eventually) thousands
of processes that all need to do this with each other
- Doesn't allow for some future use of the interface we would like to
consider adding in the future (see below)
- Interestingly reading from /proc/pid/mem currently actually
involves two copies! (But this could be fixed pretty easily)
As mentioned previously use of vmsplice instead was considered, but has
problems. Since you need the reader and writer working co-operatively if
the pipe is not drained then you block. Which requires some wrapping to
do non blocking on the send side or polling on the receive. In all to all
communication it requires ordering otherwise you can deadlock. And in the
example of many MPI tasks writing to one MPI task vmsplice serialises the
copying.
There are some cases of MPI collectives where even a single copy interface
does not get us the performance gain we could. For example in an
MPI_Reduce rather than copy the data from the source we would like to
instead use it directly in a mathops (say the reduce is doing a sum) as
this would save us doing a copy. We don't need to keep a copy of the data
from the source. I haven't implemented this, but I think this interface
could in the future do all this through the use of the flags - eg could
specify the math operation and type and the kernel rather than just
copying the data would apply the specified operation between the source
and destination and store it in the destination.
Although we don't have a "second user" of the interface (though I've had
some nibbles from people who may be interested in using it for intra
process messaging which is not MPI). This interface is something which
hardware vendors are already doing for their custom drivers to implement
fast local communication. And so in addition to this being useful for
OpenMPI it would mean the driver maintainers don't have to fix things up
when the mm changes.
There was some discussion about how much faster a true zero copy would
go. Here's a link back to the email with some testing I did on that:
http://marc.info/?l=linux-mm&m=130105930902915&w=2
There is a basic man page for the proposed interface here:
http://ozlabs.org/~cyeoh/cma/process_vm_readv.txt
This has been implemented for x86 and powerpc, other architecture should
mainly (I think) just need to add syscall numbers for the process_vm_readv
and process_vm_writev. There are 32 bit compatibility versions for
64-bit kernels.
For arch maintainers there are some simple tests to be able to quickly
verify that the syscalls are working correctly here:
http://ozlabs.org/~cyeoh/cma/cma-test-20110718.tgz
Signed-off-by: Chris Yeoh <yeohc@au1.ibm.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: David Howells <dhowells@redhat.com>
Cc: James Morris <jmorris@namei.org>
Cc: <linux-man@vger.kernel.org>
Cc: <linux-arch@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---
arch/powerpc/include/asm/systbl.h | 2
arch/powerpc/include/asm/unistd.h | 4
arch/x86/ia32/ia32entry.S | 2
arch/x86/include/asm/unistd_32.h | 4
arch/x86/include/asm/unistd_64.h | 4
arch/x86/kernel/syscall_table_32.S | 2
fs/aio.c | 4
fs/compat.c | 7
fs/read_write.c | 8
include/linux/compat.h | 3
include/linux/fs.h | 7
include/linux/syscalls.h | 13
kernel/sys_ni.c | 4
mm/Makefile | 3
mm/process_vm_access.c | 496 +++++++++++++++++++++++++++
security/keys/compat.c | 2
security/keys/keyctl.c | 2
17 files changed, 550 insertions(+), 17 deletions(-)
diff -puN arch/powerpc/include/asm/systbl.h~cross-memory-attach-v3 arch/powerpc/include/asm/systbl.h
--- a/arch/powerpc/include/asm/systbl.h~cross-memory-attach-v3
+++ a/arch/powerpc/include/asm/systbl.h
@@ -354,3 +354,5 @@ COMPAT_SYS_SPU(clock_adjtime)
SYSCALL_SPU(syncfs)
COMPAT_SYS_SPU(sendmmsg)
SYSCALL_SPU(setns)
+COMPAT_SYS(process_vm_readv)
+COMPAT_SYS(process_vm_writev)
diff -puN arch/powerpc/include/asm/unistd.h~cross-memory-attach-v3 arch/powerpc/include/asm/unistd.h
--- a/arch/powerpc/include/asm/unistd.h~cross-memory-attach-v3
+++ a/arch/powerpc/include/asm/unistd.h
@@ -373,10 +373,12 @@
#define __NR_syncfs 348
#define __NR_sendmmsg 349
#define __NR_setns 350
+#define __NR_process_vm_readv 351
+#define __NR_process_vm_writev 352
#ifdef __KERNEL__
-#define __NR_syscalls 351
+#define __NR_syscalls 353
#define __NR__exit __NR_exit
#define NR_syscalls __NR_syscalls
diff -puN arch/x86/ia32/ia32entry.S~cross-memory-attach-v3 arch/x86/ia32/ia32entry.S
--- a/arch/x86/ia32/ia32entry.S~cross-memory-attach-v3
+++ a/arch/x86/ia32/ia32entry.S
@@ -850,4 +850,6 @@ ia32_sys_call_table:
.quad sys_syncfs
.quad compat_sys_sendmmsg /* 345 */
.quad sys_setns
+ .quad compat_sys_process_vm_readv
+ .quad compat_sys_process_vm_writev
ia32_syscall_end:
diff -puN arch/x86/include/asm/unistd_32.h~cross-memory-attach-v3 arch/x86/include/asm/unistd_32.h
--- a/arch/x86/include/asm/unistd_32.h~cross-memory-attach-v3
+++ a/arch/x86/include/asm/unistd_32.h
@@ -352,10 +352,12 @@
#define __NR_syncfs 344
#define __NR_sendmmsg 345
#define __NR_setns 346
+#define __NR_process_vm_readv 347
+#define __NR_process_vm_writev 348
#ifdef __KERNEL__
-#define NR_syscalls 347
+#define NR_syscalls 349
#define __ARCH_WANT_IPC_PARSE_VERSION
#define __ARCH_WANT_OLD_READDIR
diff -puN arch/x86/include/asm/unistd_64.h~cross-memory-attach-v3 arch/x86/include/asm/unistd_64.h
--- a/arch/x86/include/asm/unistd_64.h~cross-memory-attach-v3
+++ a/arch/x86/include/asm/unistd_64.h
@@ -682,6 +682,10 @@ __SYSCALL(__NR_sendmmsg, sys_sendmmsg)
__SYSCALL(__NR_setns, sys_setns)
#define __NR_getcpu 309
__SYSCALL(__NR_getcpu, sys_getcpu)
+#define __NR_process_vm_readv 310
+__SYSCALL(__NR_process_vm_readv, sys_process_vm_readv)
+#define __NR_process_vm_writev 311
+__SYSCALL(__NR_process_vm_writev, sys_process_vm_writev)
#ifndef __NO_STUBS
#define __ARCH_WANT_OLD_READDIR
diff -puN arch/x86/kernel/syscall_table_32.S~cross-memory-attach-v3 arch/x86/kernel/syscall_table_32.S
--- a/arch/x86/kernel/syscall_table_32.S~cross-memory-attach-v3
+++ a/arch/x86/kernel/syscall_table_32.S
@@ -346,3 +346,5 @@ ENTRY(sys_call_table)
.long sys_syncfs
.long sys_sendmmsg /* 345 */
.long sys_setns
+ .long sys_process_vm_readv
+ .long sys_process_vm_writev
diff -puN fs/aio.c~cross-memory-attach-v3 fs/aio.c
--- a/fs/aio.c~cross-memory-attach-v3
+++ a/fs/aio.c
@@ -1387,13 +1387,13 @@ static ssize_t aio_setup_vectored_rw(int
ret = compat_rw_copy_check_uvector(type,
(struct compat_iovec __user *)kiocb->ki_buf,
kiocb->ki_nbytes, 1, &kiocb->ki_inline_vec,
- &kiocb->ki_iovec);
+ &kiocb->ki_iovec, 1);
else
#endif
ret = rw_copy_check_uvector(type,
(struct iovec __user *)kiocb->ki_buf,
kiocb->ki_nbytes, 1, &kiocb->ki_inline_vec,
- &kiocb->ki_iovec);
+ &kiocb->ki_iovec, 1);
if (ret < 0)
goto out;
diff -puN fs/compat.c~cross-memory-attach-v3 fs/compat.c
--- a/fs/compat.c~cross-memory-attach-v3
+++ a/fs/compat.c
@@ -546,7 +546,7 @@ out:
ssize_t compat_rw_copy_check_uvector(int type,
const struct compat_iovec __user *uvector, unsigned long nr_segs,
unsigned long fast_segs, struct iovec *fast_pointer,
- struct iovec **ret_pointer)
+ struct iovec **ret_pointer, int check_access)
{
compat_ssize_t tot_len;
struct iovec *iov = *ret_pointer = fast_pointer;
@@ -593,7 +593,8 @@ ssize_t compat_rw_copy_check_uvector(int
}
if (len < 0) /* size_t not fitting in compat_ssize_t .. */
goto out;
- if (!access_ok(vrfy_dir(type), compat_ptr(buf), len)) {
+ if (check_access &&
+ !access_ok(vrfy_dir(type), compat_ptr(buf), len)) {
ret = -EFAULT;
goto out;
}
@@ -1107,7 +1108,7 @@ static ssize_t compat_do_readv_writev(in
goto out;
tot_len = compat_rw_copy_check_uvector(type, uvector, nr_segs,
- UIO_FASTIOV, iovstack, &iov);
+ UIO_FASTIOV, iovstack, &iov, 1);
if (tot_len == 0) {
ret = 0;
goto out;
diff -puN fs/read_write.c~cross-memory-attach-v3 fs/read_write.c
--- a/fs/read_write.c~cross-memory-attach-v3
+++ a/fs/read_write.c
@@ -633,7 +633,8 @@ ssize_t do_loop_readv_writev(struct file
ssize_t rw_copy_check_uvector(int type, const struct iovec __user * uvector,
unsigned long nr_segs, unsigned long fast_segs,
struct iovec *fast_pointer,
- struct iovec **ret_pointer)
+ struct iovec **ret_pointer,
+ int check_access)
{
unsigned long seg;
ssize_t ret;
@@ -689,7 +690,8 @@ ssize_t rw_copy_check_uvector(int type,
ret = -EINVAL;
goto out;
}
- if (unlikely(!access_ok(vrfy_dir(type), buf, len))) {
+ if (check_access
+ && unlikely(!access_ok(vrfy_dir(type), buf, len))) {
ret = -EFAULT;
goto out;
}
@@ -721,7 +723,7 @@ static ssize_t do_readv_writev(int type,
}
ret = rw_copy_check_uvector(type, uvector, nr_segs,
- ARRAY_SIZE(iovstack), iovstack, &iov);
+ ARRAY_SIZE(iovstack), iovstack, &iov, 1);
if (ret <= 0)
goto out;
diff -puN include/linux/compat.h~cross-memory-attach-v3 include/linux/compat.h
--- a/include/linux/compat.h~cross-memory-attach-v3
+++ a/include/linux/compat.h
@@ -547,7 +547,8 @@ extern ssize_t compat_rw_copy_check_uvec
const struct compat_iovec __user *uvector,
unsigned long nr_segs,
unsigned long fast_segs, struct iovec *fast_pointer,
- struct iovec **ret_pointer);
+ struct iovec **ret_pointer,
+ int check_access);
extern void __user *compat_alloc_user_space(unsigned long len);
diff -puN include/linux/fs.h~cross-memory-attach-v3 include/linux/fs.h
--- a/include/linux/fs.h~cross-memory-attach-v3
+++ a/include/linux/fs.h
@@ -1633,9 +1633,10 @@ struct inode_operations {
struct seq_file;
ssize_t rw_copy_check_uvector(int type, const struct iovec __user * uvector,
- unsigned long nr_segs, unsigned long fast_segs,
- struct iovec *fast_pointer,
- struct iovec **ret_pointer);
+ unsigned long nr_segs, unsigned long fast_segs,
+ struct iovec *fast_pointer,
+ struct iovec **ret_pointer,
+ int check_access);
extern ssize_t vfs_read(struct file *, char __user *, size_t, loff_t *);
extern ssize_t vfs_write(struct file *, const char __user *, size_t, loff_t *);
diff -puN include/linux/syscalls.h~cross-memory-attach-v3 include/linux/syscalls.h
--- a/include/linux/syscalls.h~cross-memory-attach-v3
+++ a/include/linux/syscalls.h
@@ -844,4 +844,17 @@ asmlinkage long sys_open_by_handle_at(in
struct file_handle __user *handle,
int flags);
asmlinkage long sys_setns(int fd, int nstype);
+asmlinkage long sys_process_vm_readv(pid_t pid,
+ const struct iovec __user *lvec,
+ unsigned long liovcnt,
+ const struct iovec __user *rvec,
+ unsigned long riovcnt,
+ unsigned long flags);
+asmlinkage long sys_process_vm_writev(pid_t pid,
+ const struct iovec __user *lvec,
+ unsigned long liovcnt,
+ const struct iovec __user *rvec,
+ unsigned long riovcnt,
+ unsigned long flags);
+
#endif
diff -puN kernel/sys_ni.c~cross-memory-attach-v3 kernel/sys_ni.c
--- a/kernel/sys_ni.c~cross-memory-attach-v3
+++ a/kernel/sys_ni.c
@@ -145,6 +145,10 @@ cond_syscall(sys_io_submit);
cond_syscall(sys_io_cancel);
cond_syscall(sys_io_getevents);
cond_syscall(sys_syslog);
+cond_syscall(sys_process_vm_readv);
+cond_syscall(sys_process_vm_writev);
+cond_syscall(compat_sys_process_vm_readv);
+cond_syscall(compat_sys_process_vm_writev);
/* arch-specific weak syscall entries */
cond_syscall(sys_pciconfig_read);
diff -puN mm/Makefile~cross-memory-attach-v3 mm/Makefile
--- a/mm/Makefile~cross-memory-attach-v3
+++ a/mm/Makefile
@@ -5,7 +5,8 @@
mmu-y := nommu.o
mmu-$(CONFIG_MMU) := fremap.o highmem.o madvise.o memory.o mincore.o \
mlock.o mmap.o mprotect.o mremap.o msync.o rmap.o \
- vmalloc.o pagewalk.o pgtable-generic.o
+ vmalloc.o pagewalk.o pgtable-generic.o \
+ process_vm_access.o
obj-y := filemap.o mempool.o oom_kill.o fadvise.o \
maccess.o page_alloc.o page-writeback.o \
diff -puN /dev/null mm/process_vm_access.c
--- /dev/null
+++ a/mm/process_vm_access.c
@@ -0,0 +1,496 @@
+/*
+ * linux/mm/process_vm_access.c
+ *
+ * Copyright (C) 2010-2011 Christopher Yeoh <cyeoh@au1.ibm.com>, IBM Corp.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+
+#include <linux/mm.h>
+#include <linux/uio.h>
+#include <linux/sched.h>
+#include <linux/highmem.h>
+#include <linux/ptrace.h>
+#include <linux/slab.h>
+#include <linux/syscalls.h>
+
+#ifdef CONFIG_COMPAT
+#include <linux/compat.h>
+#endif
+
+/**
+ * process_vm_rw_pages - read/write pages from task specified
+ * @task: task to read/write from
+ * @mm: mm for task
+ * @process_pages: struct pages area that can store at least
+ * nr_pages_to_copy struct page pointers
+ * @pa: address of page in task to start copying from/to
+ * @start_offset: offset in page to start copying from/to
+ * @len: number of bytes to copy
+ * @lvec: iovec array specifying where to copy to/from
+ * @lvec_cnt: number of elements in iovec array
+ * @lvec_current: index in iovec array we are up to
+ * @lvec_offset: offset in bytes from current iovec iov_base we are up to
+ * @vm_write: 0 means copy from, 1 means copy to
+ * @nr_pages_to_copy: number of pages to copy
+ * @bytes_copied: returns number of bytes successfully copied
+ * Returns 0 on success, error code otherwise
+ */
+static int process_vm_rw_pages(struct task_struct *task,
+ struct mm_struct *mm,
+ struct page **process_pages,
+ unsigned long pa,
+ unsigned long start_offset,
+ unsigned long len,
+ const struct iovec *lvec,
+ unsigned long lvec_cnt,
+ unsigned long *lvec_current,
+ size_t *lvec_offset,
+ int vm_write,
+ unsigned int nr_pages_to_copy,
+ ssize_t *bytes_copied)
+{
+ int pages_pinned;
+ void *target_kaddr;
+ int pgs_copied = 0;
+ int j;
+ int ret;
+ ssize_t bytes_to_copy;
+ ssize_t rc = 0;
+
+ *bytes_copied = 0;
+
+ /* Get the pages we're interested in */
+ down_read(&mm->mmap_sem);
+ pages_pinned = get_user_pages(task, mm, pa,
+ nr_pages_to_copy,
+ vm_write, 0, process_pages, NULL);
+ up_read(&mm->mmap_sem);
+
+ if (pages_pinned != nr_pages_to_copy) {
+ rc = -EFAULT;
+ goto end;
+ }
+
+ /* Do the copy for each page */
+ for (pgs_copied = 0;
+ (pgs_copied < nr_pages_to_copy) && (*lvec_current < lvec_cnt);
+ pgs_copied++) {
+ /* Make sure we have a non zero length iovec */
+ while (*lvec_current < lvec_cnt
+ && lvec[*lvec_current].iov_len == 0)
+ (*lvec_current)++;
+ if (*lvec_current == lvec_cnt)
+ break;
+
+ /*
+ * Will copy smallest of:
+ * - bytes remaining in page
+ * - bytes remaining in destination iovec
+ */
+ bytes_to_copy = min_t(ssize_t, PAGE_SIZE - start_offset,
+ len - *bytes_copied);
+ bytes_to_copy = min_t(ssize_t, bytes_to_copy,
+ lvec[*lvec_current].iov_len
+ - *lvec_offset);
+
+ target_kaddr = kmap(process_pages[pgs_copied]) + start_offset;
+
+ if (vm_write)
+ ret = copy_from_user(target_kaddr,
+ lvec[*lvec_current].iov_base
+ + *lvec_offset,
+ bytes_to_copy);
+ else
+ ret = copy_to_user(lvec[*lvec_current].iov_base
+ + *lvec_offset,
+ target_kaddr, bytes_to_copy);
+ kunmap(process_pages[pgs_copied]);
+ if (ret) {
+ *bytes_copied += bytes_to_copy - ret;
+ pgs_copied++;
+ rc = -EFAULT;
+ goto end;
+ }
+ *bytes_copied += bytes_to_copy;
+ *lvec_offset += bytes_to_copy;
+ if (*lvec_offset == lvec[*lvec_current].iov_len) {
+ /*
+ * Need to copy remaining part of page into the
+ * next iovec if there are any bytes left in page
+ */
+ (*lvec_current)++;
+ *lvec_offset = 0;
+ start_offset = (start_offset + bytes_to_copy)
+ % PAGE_SIZE;
+ if (start_offset)
+ pgs_copied--;
+ } else {
+ start_offset = 0;
+ }
+ }
+
+end:
+ if (vm_write) {
+ for (j = 0; j < pages_pinned; j++) {
+ if (j < pgs_copied)
+ set_page_dirty_lock(process_pages[j]);
+ put_page(process_pages[j]);
+ }
+ } else {
+ for (j = 0; j < pages_pinned; j++)
+ put_page(process_pages[j]);
+ }
+
+ return rc;
+}
+
+/* Maximum number of pages kmalloc'd to hold struct page's during copy */
+#define PVM_MAX_KMALLOC_PAGES (PAGE_SIZE * 2)
+
+/**
+ * process_vm_rw_single_vec - read/write pages from task specified
+ * @addr: start memory address of target process
+ * @len: size of area to copy to/from
+ * @lvec: iovec array specifying where to copy to/from locally
+ * @lvec_cnt: number of elements in iovec array
+ * @lvec_current: index in iovec array we are up to
+ * @lvec_offset: offset in bytes from current iovec iov_base we are up to
+ * @process_pages: struct pages area that can store at least
+ * nr_pages_to_copy struct page pointers
+ * @mm: mm for task
+ * @task: task to read/write from
+ * @vm_write: 0 means copy from, 1 means copy to
+ * @bytes_copied: returns number of bytes successfully copied
+ * Returns 0 on success or on failure error code
+ */
+static int process_vm_rw_single_vec(unsigned long addr,
+ unsigned long len,
+ const struct iovec *lvec,
+ unsigned long lvec_cnt,
+ unsigned long *lvec_current,
+ size_t *lvec_offset,
+ struct page **process_pages,
+ struct mm_struct *mm,
+ struct task_struct *task,
+ int vm_write,
+ ssize_t *bytes_copied)
+{
+ unsigned long pa = addr & PAGE_MASK;
+ unsigned long start_offset = addr - pa;
+ unsigned long nr_pages;
+ ssize_t bytes_copied_loop;
+ ssize_t rc = 0;
+ unsigned long nr_pages_copied = 0;
+ unsigned long nr_pages_to_copy;
+ unsigned long max_pages_per_loop = PVM_MAX_KMALLOC_PAGES
+ / sizeof(struct pages *);
+
+ *bytes_copied = 0;
+
+ /* Work out address and page range required */
+ if (len == 0)
+ return 0;
+ nr_pages = (addr + len - 1) / PAGE_SIZE - addr / PAGE_SIZE + 1;
+
+ while ((nr_pages_copied < nr_pages) && (*lvec_current < lvec_cnt)) {
+ nr_pages_to_copy = min(nr_pages - nr_pages_copied,
+ max_pages_per_loop);
+
+ rc = process_vm_rw_pages(task, mm, process_pages, pa,
+ start_offset, len,
+ lvec, lvec_cnt,
+ lvec_current, lvec_offset,
+ vm_write, nr_pages_to_copy,
+ &bytes_copied_loop);
+ start_offset = 0;
+ *bytes_copied += bytes_copied_loop;
+
+ if (rc < 0) {
+ return rc;
+ } else {
+ len -= bytes_copied_loop;
+ nr_pages_copied += nr_pages_to_copy;
+ pa += nr_pages_to_copy * PAGE_SIZE;
+ }
+ }
+
+ return rc;
+}
+
+/* Maximum number of entries for process pages array
+ which lives on stack */
+#define PVM_MAX_PP_ARRAY_COUNT 16
+
+/**
+ * process_vm_rw_core - core of reading/writing pages from task specified
+ * @pid: PID of process to read/write from/to
+ * @lvec: iovec array specifying where to copy to/from locally
+ * @liovcnt: size of lvec array
+ * @rvec: iovec array specifying where to copy to/from in the other process
+ * @riovcnt: size of rvec array
+ * @flags: currently unused
+ * @vm_write: 0 if reading from other process, 1 if writing to other process
+ * Returns the number of bytes read/written or error code. May
+ * return less bytes than expected if an error occurs during the copying
+ * process.
+ */
+static ssize_t process_vm_rw_core(pid_t pid, const struct iovec *lvec,
+ unsigned long liovcnt,
+ const struct iovec *rvec,
+ unsigned long riovcnt,
+ unsigned long flags, int vm_write)
+{
+ struct task_struct *task;
+ struct page *pp_stack[PVM_MAX_PP_ARRAY_COUNT];
+ struct page **process_pages = pp_stack;
+ struct mm_struct *mm;
+ unsigned long i;
+ ssize_t rc = 0;
+ ssize_t bytes_copied_loop;
+ ssize_t bytes_copied = 0;
+ unsigned long nr_pages = 0;
+ unsigned long nr_pages_iov;
+ unsigned long iov_l_curr_idx = 0;
+ size_t iov_l_curr_offset = 0;
+ ssize_t iov_len;
+
+ /*
+ * Work out how many pages of struct pages we're going to need
+ * when eventually calling get_user_pages
+ */
+ for (i = 0; i < riovcnt; i++) {
+ iov_len = rvec[i].iov_len;
+ if (iov_len > 0) {
+ nr_pages_iov = ((unsigned long)rvec[i].iov_base
+ + iov_len)
+ / PAGE_SIZE - (unsigned long)rvec[i].iov_base
+ / PAGE_SIZE + 1;
+ nr_pages = max(nr_pages, nr_pages_iov);
+ }
+ }
+
+ if (nr_pages == 0)
+ return 0;
+
+ if (nr_pages > PVM_MAX_PP_ARRAY_COUNT) {
+ /* For reliability don't try to kmalloc more than
+ 2 pages worth */
+ process_pages = kmalloc(min_t(size_t, PVM_MAX_KMALLOC_PAGES,
+ sizeof(struct pages *)*nr_pages),
+ GFP_KERNEL);
+
+ if (!process_pages)
+ return -ENOMEM;
+ }
+
+ /* Get process information */
+ rcu_read_lock();
+ task = find_task_by_vpid(pid);
+ if (task)
+ get_task_struct(task);
+ rcu_read_unlock();
+ if (!task) {
+ rc = -ESRCH;
+ goto free_proc_pages;
+ }
+
+ task_lock(task);
+ if (__ptrace_may_access(task, PTRACE_MODE_ATTACH)) {
+ task_unlock(task);
+ rc = -EPERM;
+ goto put_task_struct;
+ }
+ mm = task->mm;
+
+ if (!mm || (task->flags & PF_KTHREAD)) {
+ task_unlock(task);
+ rc = -EINVAL;
+ goto put_task_struct;
+ }
+
+ atomic_inc(&mm->mm_users);
+ task_unlock(task);
+
+ for (i = 0; i < riovcnt && iov_l_curr_idx < liovcnt; i++) {
+ rc = process_vm_rw_single_vec(
+ (unsigned long)rvec[i].iov_base, rvec[i].iov_len,
+ lvec, liovcnt, &iov_l_curr_idx, &iov_l_curr_offset,
+ process_pages, mm, task, vm_write, &bytes_copied_loop);
+ bytes_copied += bytes_copied_loop;
+ if (rc != 0) {
+ /* If we have managed to copy any data at all then
+ we return the number of bytes copied. Otherwise
+ we return the error code */
+ if (bytes_copied)
+ rc = bytes_copied;
+ goto put_mm;
+ }
+ }
+
+ rc = bytes_copied;
+put_mm:
+ mmput(mm);
+
+put_task_struct:
+ put_task_struct(task);
+
+free_proc_pages:
+ if (process_pages != pp_stack)
+ kfree(process_pages);
+ return rc;
+}
+
+/**
+ * process_vm_rw - check iovecs before calling core routine
+ * @pid: PID of process to read/write from/to
+ * @lvec: iovec array specifying where to copy to/from locally
+ * @liovcnt: size of lvec array
+ * @rvec: iovec array specifying where to copy to/from in the other process
+ * @riovcnt: size of rvec array
+ * @flags: currently unused
+ * @vm_write: 0 if reading from other process, 1 if writing to other process
+ * Returns the number of bytes read/written or error code. May
+ * return less bytes than expected if an error occurs during the copying
+ * process.
+ */
+static ssize_t process_vm_rw(pid_t pid,
+ const struct iovec __user *lvec,
+ unsigned long liovcnt,
+ const struct iovec __user *rvec,
+ unsigned long riovcnt,
+ unsigned long flags, int vm_write)
+{
+ struct iovec iovstack_l[UIO_FASTIOV];
+ struct iovec iovstack_r[UIO_FASTIOV];
+ struct iovec *iov_l = iovstack_l;
+ struct iovec *iov_r = iovstack_r;
+ ssize_t rc;
+
+ if (flags != 0)
+ return -EINVAL;
+
+ /* Check iovecs */
+ if (vm_write)
+ rc = rw_copy_check_uvector(WRITE, lvec, liovcnt, UIO_FASTIOV,
+ iovstack_l, &iov_l, 1);
+ else
+ rc = rw_copy_check_uvector(READ, lvec, liovcnt, UIO_FASTIOV,
+ iovstack_l, &iov_l, 1);
+ if (rc <= 0)
+ goto free_iovecs;
+
+ rc = rw_copy_check_uvector(READ, rvec, riovcnt, UIO_FASTIOV,
+ iovstack_r, &iov_r, 0);
+ if (rc <= 0)
+ goto free_iovecs;
+
+ rc = process_vm_rw_core(pid, iov_l, liovcnt, iov_r, riovcnt, flags,
+ vm_write);
+
+free_iovecs:
+ if (iov_r != iovstack_r)
+ kfree(iov_r);
+ if (iov_l != iovstack_l)
+ kfree(iov_l);
+
+ return rc;
+}
+
+SYSCALL_DEFINE6(process_vm_readv, pid_t, pid, const struct iovec __user *, lvec,
+ unsigned long, liovcnt, const struct iovec __user *, rvec,
+ unsigned long, riovcnt, unsigned long, flags)
+{
+ return process_vm_rw(pid, lvec, liovcnt, rvec, riovcnt, flags, 0);
+}
+
+SYSCALL_DEFINE6(process_vm_writev, pid_t, pid,
+ const struct iovec __user *, lvec,
+ unsigned long, liovcnt, const struct iovec __user *, rvec,
+ unsigned long, riovcnt, unsigned long, flags)
+{
+ return process_vm_rw(pid, lvec, liovcnt, rvec, riovcnt, flags, 1);
+}
+
+#ifdef CONFIG_COMPAT
+
+asmlinkage ssize_t
+compat_process_vm_rw(compat_pid_t pid,
+ const struct compat_iovec __user *lvec,
+ unsigned long liovcnt,
+ const struct compat_iovec __user *rvec,
+ unsigned long riovcnt,
+ unsigned long flags, int vm_write)
+{
+ struct iovec iovstack_l[UIO_FASTIOV];
+ struct iovec iovstack_r[UIO_FASTIOV];
+ struct iovec *iov_l = iovstack_l;
+ struct iovec *iov_r = iovstack_r;
+ ssize_t rc = -EFAULT;
+
+ if (flags != 0)
+ return -EINVAL;
+
+ if (!access_ok(VERIFY_READ, lvec, liovcnt * sizeof(*lvec)))
+ goto out;
+
+ if (!access_ok(VERIFY_READ, rvec, riovcnt * sizeof(*rvec)))
+ goto out;
+
+ if (vm_write)
+ rc = compat_rw_copy_check_uvector(WRITE, lvec, liovcnt,
+ UIO_FASTIOV, iovstack_l,
+ &iov_l, 1);
+ else
+ rc = compat_rw_copy_check_uvector(READ, lvec, liovcnt,
+ UIO_FASTIOV, iovstack_l,
+ &iov_l, 1);
+ if (rc <= 0)
+ goto free_iovecs;
+ rc = compat_rw_copy_check_uvector(READ, rvec, riovcnt,
+ UIO_FASTIOV, iovstack_r,
+ &iov_r, 0);
+ if (rc <= 0)
+ goto free_iovecs;
+
+ rc = process_vm_rw_core(pid, iov_l, liovcnt, iov_r, riovcnt, flags,
+ vm_write);
+
+free_iovecs:
+ if (iov_r != iovstack_r)
+ kfree(iov_r);
+ if (iov_l != iovstack_l)
+ kfree(iov_l);
+
+out:
+ return rc;
+}
+
+asmlinkage ssize_t
+compat_sys_process_vm_readv(compat_pid_t pid,
+ const struct compat_iovec __user *lvec,
+ unsigned long liovcnt,
+ const struct compat_iovec __user *rvec,
+ unsigned long riovcnt,
+ unsigned long flags)
+{
+ return compat_process_vm_rw(pid, lvec, liovcnt, rvec,
+ riovcnt, flags, 0);
+}
+
+asmlinkage ssize_t
+compat_sys_process_vm_writev(compat_pid_t pid,
+ const struct compat_iovec __user *lvec,
+ unsigned long liovcnt,
+ const struct compat_iovec __user *rvec,
+ unsigned long riovcnt,
+ unsigned long flags)
+{
+ return compat_process_vm_rw(pid, lvec, liovcnt, rvec,
+ riovcnt, flags, 1);
+}
+
+#endif
diff -puN security/keys/compat.c~cross-memory-attach-v3 security/keys/compat.c
--- a/security/keys/compat.c~cross-memory-attach-v3
+++ a/security/keys/compat.c
@@ -38,7 +38,7 @@ long compat_keyctl_instantiate_key_iov(
ret = compat_rw_copy_check_uvector(WRITE, _payload_iov, ioc,
ARRAY_SIZE(iovstack),
- iovstack, &iov);
+ iovstack, &iov, 1);
if (ret < 0)
return ret;
if (ret == 0)
diff -puN security/keys/keyctl.c~cross-memory-attach-v3 security/keys/keyctl.c
--- a/security/keys/keyctl.c~cross-memory-attach-v3
+++ a/security/keys/keyctl.c
@@ -1065,7 +1065,7 @@ long keyctl_instantiate_key_iov(key_seri
goto no_payload;
ret = rw_copy_check_uvector(WRITE, _payload_iov, ioc,
- ARRAY_SIZE(iovstack), iovstack, &iov);
+ ARRAY_SIZE(iovstack), iovstack, &iov, 1);
if (ret < 0)
return ret;
if (ret == 0)
_
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.