Add the xstate regset support which helps extend the kernel ptrace and the core-dump interfaces to support AVX state etc. This regset interface is designed to support all the future state that gets supported using xsave/xrstor infrastructure. Looking at the memory layout saved by "xsave", one can't say which state is represented in the memory layout. This is because if a particular state is in init state, in the xsave hdr it can be represented by bit '0'. And hence we can't really say by the xsave header wether a state is in init state or the state is not saved in the memory layout. And hence the xsave memory layout available through this regset interface uses SW usable bytes [464..511] to convey what state is represented in the memory layout. First 8 bytes of the sw_usable_bytes[464..467] will be set to OS enabled xstate mask(which is same as the 64bit mask returned by the xgetbv's xCR0). The note NT_X86_XSTATE represents the extended state information in the core file, using the above mentioned memory layout. Signed-off-by: Suresh Siddha Signed-off-by: Hongjiu Lu --- arch/x86/include/asm/i387.h | 12 +++++- arch/x86/include/asm/user.h | 58 ++++++++++++++++++++++++++++++ arch/x86/include/asm/xsave.h | 2 + arch/x86/kernel/i387.c | 83 +++++++++++++++++++++++++++++++++++++++++++ arch/x86/kernel/ptrace.c | 34 ++++++++++++++++- arch/x86/kernel/xsave.c | 1 include/linux/elf.h | 1 7 files changed, 187 insertions(+), 4 deletions(-) Index: tip/arch/x86/include/asm/i387.h =================================================================== --- tip.orig/arch/x86/include/asm/i387.h +++ tip/arch/x86/include/asm/i387.h @@ -33,8 +33,16 @@ extern void init_thread_xstate(void); extern int dump_fpu(struct pt_regs *, struct user_i387_struct *); extern user_regset_active_fn fpregs_active, xfpregs_active; -extern user_regset_get_fn fpregs_get, xfpregs_get, fpregs_soft_get; -extern user_regset_set_fn fpregs_set, xfpregs_set, fpregs_soft_set; +extern user_regset_get_fn fpregs_get, xfpregs_get, fpregs_soft_get, + xstateregs_get; +extern user_regset_set_fn fpregs_set, xfpregs_set, fpregs_soft_set, + xstateregs_set; + +/* + * xstateregs_active == fpregs_active. Please refer to the comment + * at the definition of fpregs_active. + */ +#define xstateregs_active fpregs_active extern struct _fpx_sw_bytes fx_sw_reserved; #ifdef CONFIG_IA32_EMULATION Index: tip/arch/x86/include/asm/xsave.h =================================================================== --- tip.orig/arch/x86/include/asm/xsave.h +++ tip/arch/x86/include/asm/xsave.h @@ -27,9 +27,11 @@ extern unsigned int xstate_size; extern u64 pcntxt_mask; extern struct xsave_struct *init_xstate_buf; +extern u64 xstate_fx_sw_bytes[USER_XSTATE_FX_SW_WORDS]; extern void xsave_cntxt_init(void); extern void xsave_init(void); +extern void update_regset_xstate_info(unsigned int size, u64 xstate_mask); extern int init_fpu(struct task_struct *child); extern int check_for_xstate(struct i387_fxsave_struct __user *buf, void __user *fpstate, Index: tip/arch/x86/kernel/i387.c =================================================================== --- tip.orig/arch/x86/kernel/i387.c +++ tip/arch/x86/kernel/i387.c @@ -164,6 +164,11 @@ int init_fpu(struct task_struct *tsk) return 0; } +/* + * The xstateregs_active() routine is the same as the fpregs_active() routine, + * as the "regset->n" for the xstate regset will be updated based on the feature + * capabilites supported by the xsave. + */ int fpregs_active(struct task_struct *target, const struct user_regset *regset) { return tsk_used_math(target) ? regset->n : 0; @@ -224,6 +229,84 @@ int xfpregs_set(struct task_struct *targ return ret; } +int xstateregs_get(struct task_struct *target, const struct user_regset *regset, + unsigned int pos, unsigned int count, + void *kbuf, void __user *ubuf) +{ + int ret; + + if (!cpu_has_xsave) + return -ENODEV; + + ret = init_fpu(target); + if (ret) + return ret; + + /* + * First copy the fxsave bytes 0..463. + */ + ret = user_regset_copyout(&pos, &count, &kbuf, &ubuf, + &target->thread.xstate->xsave, 0, + offsetof(struct user_xstateregs, + i387.xstate_fx_sw)); + if (ret) + return ret; + + /* + * Copy the 48bytes defined by software. + */ + ret = user_regset_copyout(&pos, &count, &kbuf, &ubuf, + xstate_fx_sw_bytes, + offsetof(struct user_xstateregs, + i387.xstate_fx_sw), + offsetof(struct user_xstateregs, + xsave_hdr)); + if (ret) + return ret; + + /* + * Copy the rest of xstate memory layout. + */ + ret = user_regset_copyout(&pos, &count, &kbuf, &ubuf, + &target->thread.xstate->xsave.xsave_hdr, + offsetof(struct user_xstateregs, + xsave_hdr), -1); + return ret; +} + +int xstateregs_set(struct task_struct *target, const struct user_regset *regset, + unsigned int pos, unsigned int count, + const void *kbuf, const void __user *ubuf) +{ + int ret; + struct xsave_hdr_struct *xsave_hdr; + + if (!cpu_has_xsave) + return -ENODEV; + + ret = init_fpu(target); + if (ret) + return ret; + + ret = user_regset_copyin(&pos, &count, &kbuf, &ubuf, + &target->thread.xstate->xsave, 0, -1); + + /* + * mxcsr reserved bits must be masked to zero for security reasons. + */ + target->thread.xstate->fxsave.mxcsr &= mxcsr_feature_mask; + + xsave_hdr = &target->thread.xstate->xsave.xsave_hdr; + + xsave_hdr->xstate_bv &= pcntxt_mask; + /* + * These bits must be zero. + */ + xsave_hdr->reserved1[0] = xsave_hdr->reserved1[1] = 0; + + return ret; +} + #if defined CONFIG_X86_32 || defined CONFIG_IA32_EMULATION /* Index: tip/arch/x86/kernel/ptrace.c =================================================================== --- tip.orig/arch/x86/kernel/ptrace.c +++ tip/arch/x86/kernel/ptrace.c @@ -48,6 +48,7 @@ enum x86_regset { REGSET_FP, REGSET_XFP, REGSET_IOPERM64 = REGSET_XFP, + REGSET_XSTATE, REGSET_TLS, REGSET_IOPERM32, }; @@ -1560,7 +1561,7 @@ long compat_arch_ptrace(struct task_stru #ifdef CONFIG_X86_64 -static const struct user_regset x86_64_regsets[] = { +static struct user_regset x86_64_regsets[] __read_mostly = { [REGSET_GENERAL] = { .core_note_type = NT_PRSTATUS, .n = sizeof(struct user_regs_struct) / sizeof(long), @@ -1573,6 +1574,12 @@ static const struct user_regset x86_64_r .size = sizeof(long), .align = sizeof(long), .active = xfpregs_active, .get = xfpregs_get, .set = xfpregs_set }, + [REGSET_XSTATE] = { + .core_note_type = NT_X86_XSTATE, + .size = sizeof(u64), .align = sizeof(u64), + .active = xstateregs_active, .get = xstateregs_get, + .set = xstateregs_set + }, [REGSET_IOPERM64] = { .core_note_type = NT_386_IOPERM, .n = IO_BITMAP_LONGS, @@ -1598,7 +1605,7 @@ static const struct user_regset_view use #endif /* CONFIG_X86_64 */ #if defined CONFIG_X86_32 || defined CONFIG_IA32_EMULATION -static const struct user_regset x86_32_regsets[] = { +static struct user_regset x86_32_regsets[] __read_mostly = { [REGSET_GENERAL] = { .core_note_type = NT_PRSTATUS, .n = sizeof(struct user_regs_struct32) / sizeof(u32), @@ -1617,6 +1624,12 @@ static const struct user_regset x86_32_r .size = sizeof(u32), .align = sizeof(u32), .active = xfpregs_active, .get = xfpregs_get, .set = xfpregs_set }, + [REGSET_XSTATE] = { + .core_note_type = NT_X86_XSTATE, + .size = sizeof(u64), .align = sizeof(u64), + .active = xstateregs_active, .get = xstateregs_get, + .set = xstateregs_set + }, [REGSET_TLS] = { .core_note_type = NT_386_TLS, .n = GDT_ENTRY_TLS_ENTRIES, .bias = GDT_ENTRY_TLS_MIN, @@ -1639,6 +1652,23 @@ static const struct user_regset_view use }; #endif +/* + * This represents bytes 464..511 in the memory layout exported through + * the REGSET_XSTATE interface. + */ +u64 xstate_fx_sw_bytes[USER_XSTATE_FX_SW_WORDS]; + +void update_regset_xstate_info(unsigned int size, u64 xstate_mask) +{ +#ifdef CONFIG_X86_64 + x86_64_regsets[REGSET_XSTATE].n = size / sizeof(u64); +#endif +#if defined CONFIG_X86_32 || defined CONFIG_IA32_EMULATION + x86_32_regsets[REGSET_XSTATE].n = size / sizeof(u64); +#endif + xstate_fx_sw_bytes[USER_XSTATE_XCR0_WORD] = xstate_mask; +} + const struct user_regset_view *task_user_regset_view(struct task_struct *task) { #ifdef CONFIG_IA32_EMULATION Index: tip/arch/x86/kernel/xsave.c =================================================================== --- tip.orig/arch/x86/kernel/xsave.c +++ tip/arch/x86/kernel/xsave.c @@ -337,6 +337,7 @@ void __ref xsave_cntxt_init(void) cpuid_count(0xd, 0, &eax, &ebx, &ecx, &edx); xstate_size = ebx; + update_regset_xstate_info(xstate_size, pcntxt_mask); prepare_fx_sw_frame(); setup_xstate_init(); Index: tip/include/linux/elf.h =================================================================== --- tip.orig/include/linux/elf.h +++ tip/include/linux/elf.h @@ -361,6 +361,7 @@ typedef struct elf64_shdr { #define NT_PPC_VSX 0x102 /* PowerPC VSX registers */ #define NT_386_TLS 0x200 /* i386 TLS slots (struct user_desc) */ #define NT_386_IOPERM 0x201 /* x86 io permission bitmap (1=deny) */ +#define NT_X86_XSTATE 0x202 /* x86 extended state using xsave */ #define NT_S390_HIGH_GPRS 0x300 /* s390 upper register halves */ Index: tip/arch/x86/include/asm/user.h =================================================================== --- tip.orig/arch/x86/include/asm/user.h +++ tip/arch/x86/include/asm/user.h @@ -1,5 +1,63 @@ +#ifndef _ASM_X86_USER_H +#define _ASM_X86_USER_H + #ifdef CONFIG_X86_32 # include "user_32.h" #else # include "user_64.h" #endif + +#include + +struct user_ymmh_regs { + /* 16 * 16 bytes for each YMMH-reg */ + __u32 ymmh_space[64]; +}; + +struct user_xsave_hdr { + __u64 xstate_bv; + __u64 reserved1[2]; + __u64 reserved2[5]; +}; + +/* + * The structure layout of user_xstateregs, used for exporting the + * extended register state through ptrace and core-dump (NT_X86_XSTATE note) + * interfaces will be same as the memory layout of xsave used by the processor + * (except for the bytes 464..511, which can be used by the software) and hence + * the size of this structure varies depending on the features supported by the + * processor and OS. The size of the structure that users need to use can be + * obtained by doing: + * cpuid_count(0xd, 0, &eax, &ptrace_xstateregs_struct_size, &ecx, &edx); + * i.e., cpuid.(eax=0xd,ecx=0).ebx will be the size that user (debuggers, etc.) + * need to use. + * + * For now, only the first 8 bytes of the software usable bytes[464..471] will + * be used and will be set to OS enabled xstate mask (which is same as the + * 64bit mask returned by the xgetbv's xCR0). Users (analyzing core dump + * remotely, etc.) can use this mask as well as the mask saved in the + * xstate_hdr bytes and interpret what states the processor/OS supports + * and what states are in modified/initialized conditions for the + * particular process/thread. + * + * Also when the user modifies certain state FP/SSE/etc through the + * ptrace interface, they must ensure that the xsave_hdr.xstate_bv + * bytes[512..519] of the memory layout are updated correspondingly. + * i.e., for example when FP state is modified to a non-init state, + * xsave_hdr.xstate_bv's bit 0 must be set to '1', when SSE is modified to + * non-init state, xsave_hdr.xstate_bv's bit 1 must to be set to '1', etc. + */ +#define USER_XSTATE_FX_SW_WORDS 6 +#define USER_XSTATE_XCR0_WORD 0 + +struct user_xstateregs { + struct { + __u64 fpx_space[58]; + __u64 xstate_fx_sw[USER_XSTATE_FX_SW_WORDS]; + } i387; + struct user_xsave_hdr xsave_hdr; + struct user_ymmh_regs ymmh; + /* further processor state extensions go here */ +}; + +#endif /* _ASM_X86_USER_H */ -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/