From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 24722 invoked by alias); 29 Mar 2010 01:07:50 -0000 Received: (qmail 24705 invoked by uid 22791); 29 Mar 2010 01:07:45 -0000 X-SWARE-Spam-Status: No, hits=0.5 required=5.0 tests=AWL,BAYES_50,NO_DNS_FOR_FROM X-Spam-Check-By: sourceware.org Received: from mga05.intel.com (HELO fmsmga101.fm.intel.com) (192.55.52.89) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Mon, 29 Mar 2010 01:07:36 +0000 Received: from fmsmga002.fm.intel.com ([10.253.24.26]) by fmsmga101.fm.intel.com with ESMTP; 28 Mar 2010 18:02:49 -0700 X-ExtLoop1: 1 Received: from gnu-6.sc.intel.com ([10.3.194.107]) by fmsmga002.fm.intel.com with ESMTP; 28 Mar 2010 18:06:12 -0700 Received: by gnu-6.sc.intel.com (Postfix, from userid 500) id 61461812387; Sun, 28 Mar 2010 18:07:34 -0700 (PDT) Date: Mon, 29 Mar 2010 01:07:00 -0000 From: "H.J. Lu" To: GDB Subject: PATCH: 4/6 [3rd try]: Add AVX support (amd64 changes) Message-ID: <20100329010734.GA27152@intel.com> Reply-To: "H.J. Lu" References: <20100304180219.GA10826@intel.com> <20100304180408.GA10869@intel.com> <20100304180748.GC10869@intel.com> <20100306222122.GE21133@intel.com> <20100307213304.GB7170@intel.com> <20100312170051.GA6321@intel.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20100312170051.GA6321@intel.com> User-Agent: Mutt/1.5.20 (2009-08-17) Mailing-List: contact gdb-patches-help@sourceware.org; run by ezmlm Precedence: bulk List-Id: List-Subscribe: List-Archive: List-Post: List-Help: , Sender: gdb-patches-owner@sourceware.org X-SW-Source: 2010-03/txt/msg00969.txt.bz2 Here are the amd64 changes to support AVX with AVX testcases. I also need to import cpuid.h from gcc 4.4 since AVX testcases need ECX from cpuid. OK to install? H.J. ---- gdb/ 2010-03-28 H.J. Lu * amd64-linux-nat.c: Include "regset.h", "elf/common.h", and "i386-xstate.h". (PTRACE_GETREGSET): New. (PTRACE_SETREGSET): Likewise. (have_ptrace_getregset): Likewise. (amd64_linux_gregset64_reg_offset): Include 16 upper YMM registers. (amd64_linux_gregset32_reg_offset): Include 8 upper YMM registers. (amd64_linux_fetch_inferior_registers): Support PTRACE_GETFPREGS. (amd64_linux_store_inferior_registers): Likewise. (amd64_linux_read_description): Check and enable AVX target descriptions. * amd64-linux-tdep.c: Include "regset.h", "i386-linux-tdep.h" and "features/i386/amd64-avx-linux.c". (amd64_linux_regset_sections): New. (amd64_linux_update_xstateregset): Likewise. (amd64_linux_core_read_description): Check and enable AVX target description. (amd64_linux_init_abi): Set xsave_xcr0_offset. Call set_gdbarch_core_regset_sections. (_initialize_amd64_linux_tdep): Call initialize_tdesc_amd64_avx_linux. * amd64-linux-tdep.h (AMD64_LINUX_ORIG_RAX_REGNUM): Replace AMD64_MXCSR_REGNUM with AMD64_YMM15H_REGNUM. (tdesc_amd64_avx_linux): New. (amd64_linux_update_xstateregset): Likewise. * amd64-tdep.c: Include "features/i386/amd64-avx.c". (amd64_ymm_names): New. (amd64_ymmh_names): Likewise. (amd64_register_name): Likewise. (amd64_supply_xstateregset): Likewise. (amd64_collect_xstateregset): Likewise. (amd64_supply_xsave): Likewise. (amd64_collect_xsave): Likewise. (AMD64_NUM_REGS): Removed. (amd64_dwarf_reg_to_regnum): Return %ymmN register number for %xmmN if AVX is available. (amd64_pseudo_register_name): Support pseudo YMM registers. (amd64_regset_from_core_section): Support .reg-xstate section. (amd64_init_abi): Set ymmh_register_names, num_ymm_regs and ymm0h_regnum. Call set_gdbarch_register_name. (amd64_init_abi): Call initialize_tdesc_amd64_avx. * amd64-tdep.h (amd64_regnum): Add AMD64_YMM0H_REGNUM and AMD64_YMM15H_REGNUM. (AMD64_NUM_REGS): New. (amd64_supply_xsave): Likewise. (amd64_collect_xsave): Likewise. (amd64_register_name): Removed. (amd64_register_type): Likewise. gdb/testsuite/ 2010-03-28 H.J. Lu * gdb.arch/i386-avx.c: New. * gdb.arch/i386-avx.exp: Likewise. * gdb.arch/i386-cpuid.h: Updated from gcc 4.4. diff --git a/gdb/amd64-linux-nat.c b/gdb/amd64-linux-nat.c index b9d5833..eb3957c 100644 --- a/gdb/amd64-linux-nat.c +++ b/gdb/amd64-linux-nat.c @@ -23,11 +23,14 @@ #include "inferior.h" #include "gdbcore.h" #include "regcache.h" +#include "regset.h" #include "linux-nat.h" #include "amd64-linux-tdep.h" #include "gdb_assert.h" #include "gdb_string.h" +#include "elf/common.h" +#include #include #include #include @@ -51,6 +54,18 @@ #include "i386-linux-tdep.h" #include "amd64-nat.h" #include "i386-nat.h" +#include "i386-xstate.h" + +#ifndef PTRACE_GETREGSET +#define PTRACE_GETREGSET 0x4204 +#endif + +#ifndef PTRACE_SETREGSET +#define PTRACE_SETREGSET 0x4205 +#endif + +/* Does the current host support PTRACE_GETREGSET? */ +static int have_ptrace_getregset = -1; /* Mapping between the general-purpose registers in GNU/Linux x86-64 `struct user' format and GDB's register cache layout. */ @@ -73,6 +88,8 @@ static int amd64_linux_gregset64_reg_offset[] = -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, + -1, -1, -1, -1, -1, -1, -1, -1, + -1, -1, -1, -1, -1, -1, -1, -1, ORIG_RAX * 8 }; @@ -99,6 +116,7 @@ static int amd64_linux_gregset32_reg_offset[] = -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, -1, + -1, -1, -1, -1, -1, -1, -1, -1, ORIG_RAX * 8 /* "orig_eax" */ }; @@ -183,10 +201,26 @@ amd64_linux_fetch_inferior_registers (struct target_ops *ops, { elf_fpregset_t fpregs; - if (ptrace (PTRACE_GETFPREGS, tid, 0, (long) &fpregs) < 0) - perror_with_name (_("Couldn't get floating point status")); + if (have_ptrace_getregset) + { + char xstateregs[I386_XSTATE_MAX_SIZE]; + struct iovec iov; - amd64_supply_fxsave (regcache, -1, &fpregs); + iov.iov_base = xstateregs; + iov.iov_len = sizeof (xstateregs); + if (ptrace (PTRACE_GETREGSET, tid, + (unsigned int) NT_X86_XSTATE, (long) &iov) < 0) + perror_with_name (_("Couldn't get extended state status")); + + amd64_supply_xsave (regcache, -1, xstateregs); + } + else + { + if (ptrace (PTRACE_GETFPREGS, tid, 0, (long) &fpregs) < 0) + perror_with_name (_("Couldn't get floating point status")); + + amd64_supply_fxsave (regcache, -1, &fpregs); + } } } @@ -226,15 +260,33 @@ amd64_linux_store_inferior_registers (struct target_ops *ops, { elf_fpregset_t fpregs; - if (ptrace (PTRACE_GETFPREGS, tid, 0, (long) &fpregs) < 0) - perror_with_name (_("Couldn't get floating point status")); + if (have_ptrace_getregset) + { + char xstateregs[I386_XSTATE_MAX_SIZE]; + struct iovec iov; + + iov.iov_base = xstateregs; + iov.iov_len = sizeof (xstateregs); + if (ptrace (PTRACE_GETREGSET, tid, + (unsigned int) NT_X86_XSTATE, (long) &iov) < 0) + perror_with_name (_("Couldn't get extended state status")); + + amd64_collect_xsave (regcache, regnum, xstateregs, 0); - amd64_collect_fxsave (regcache, regnum, &fpregs); + if (ptrace (PTRACE_SETREGSET, tid, + (unsigned int) NT_X86_XSTATE, (long) &iov) < 0) + perror_with_name (_("Couldn't write extended state status")); + } + else + { + if (ptrace (PTRACE_GETFPREGS, tid, 0, (long) &fpregs) < 0) + perror_with_name (_("Couldn't get floating point status")); - if (ptrace (PTRACE_SETFPREGS, tid, 0, (long) &fpregs) < 0) - perror_with_name (_("Couldn't write floating point status")); + amd64_collect_fxsave (regcache, regnum, &fpregs); - return; + if (ptrace (PTRACE_SETFPREGS, tid, 0, (long) &fpregs) < 0) + perror_with_name (_("Couldn't write floating point status")); + } } } @@ -688,6 +740,8 @@ amd64_linux_read_description (struct target_ops *ops) { unsigned long cs; int tid; + int is_64bit; + static uint64_t xcr0; /* GNU/Linux LWP ID's are process ID's. */ tid = TIDGET (inferior_ptid); @@ -701,10 +755,54 @@ amd64_linux_read_description (struct target_ops *ops) if (errno != 0) perror_with_name (_("Couldn't get CS register")); - if (cs == AMD64_LINUX_USER64_CS) - return tdesc_amd64_linux; + is_64bit = cs == AMD64_LINUX_USER64_CS; + + if (have_ptrace_getregset == -1) + { + uint64_t xstateregs[(I386_XSTATE_SSE_SIZE / sizeof (uint64_t))]; + struct iovec iov; + unsigned int xstate_size; + + iov.iov_base = xstateregs; + iov.iov_len = sizeof (xstateregs); + + /* Check if PTRACE_GETREGSET works. */ + if (ptrace (PTRACE_GETREGSET, tid, + (unsigned int) NT_X86_XSTATE, (long) &iov) < 0) + { + have_ptrace_getregset = 0; + xstate_size = 0; + } + else + { + have_ptrace_getregset = 1; + + /* Get XCR0 from XSAVE extended state. */ + xcr0 = xstateregs[(I386_LINUX_XSAVE_XCR0_OFFSET + / sizeof (uint64_t))]; + + xstate_size = I386_XSTATE_SIZE (xcr0); + } + + amd64_linux_update_xstateregset (xstate_size); + } + + /* Check the native XCR0 only if PTRACE_GETREGSET is available. */ + if (have_ptrace_getregset + && (xcr0 & I386_XSTATE_AVX_MASK) == I386_XSTATE_AVX_MASK) + { + if (is_64bit) + return tdesc_amd64_avx_linux; + else + return tdesc_i386_avx_linux; + } else - return tdesc_i386_linux; + { + if (is_64bit) + return tdesc_amd64_linux; + else + return tdesc_i386_linux; + } } /* Provide a prototype to silence -Wmissing-prototypes. */ diff --git a/gdb/amd64-linux-tdep.c b/gdb/amd64-linux-tdep.c index 4ad6dc9..4cc4045 100644 --- a/gdb/amd64-linux-tdep.c +++ b/gdb/amd64-linux-tdep.c @@ -28,8 +28,11 @@ #include "symtab.h" #include "gdbtypes.h" #include "reggroups.h" +#include "regset.h" #include "amd64-linux-tdep.h" +#include "i386-linux-tdep.h" #include "linux-tdep.h" +#include "i386-xstate.h" #include "gdb_string.h" @@ -38,6 +41,7 @@ #include "xml-syscall.h" #include "features/i386/amd64-linux.c" +#include "features/i386/amd64-avx-linux.c" /* The syscall's XML filename for i386. */ #define XML_SYSCALL_FILENAME_AMD64 "syscalls/amd64-linux.xml" @@ -45,6 +49,15 @@ #include "record.h" #include "linux-record.h" +/* Supported register note sections. */ +static struct core_regset_section amd64_linux_regset_sections[] = +{ + { ".reg", 144, "general-purpose" }, + { ".reg2", 512, "floating-point" }, + { ".reg-xstate", 0, "XSAVE extended state" }, + { NULL, 0 } +}; + /* Mapping between the general-purpose registers in `struct user' format and GDB's register cache layout. */ @@ -1242,6 +1255,22 @@ amd64_linux_record_signal (struct gdbarch *gdbarch, return 0; } +/* Update XSAVE extended state register note section. */ + +void +amd64_linux_update_xstateregset (unsigned int xstate_size) +{ + struct core_regset_section *xstate = &amd64_linux_regset_sections[2]; + + /* Update the XSAVE extended state register note section for "gcore". + Disable it if its size is 0. */ + gdb_assert (strcmp (xstate->sect_name, ".reg-xstate") == 0); + if (xstate_size) + xstate->size = xstate_size; + else + xstate->sect_name = NULL; +} + /* Get Linux/x86 target description from core dump. */ static const struct target_desc * @@ -1250,12 +1279,17 @@ amd64_linux_core_read_description (struct gdbarch *gdbarch, bfd *abfd) { asection *section = bfd_get_section_by_name (abfd, ".reg2"); + uint64_t xcr0; if (section == NULL) return NULL; /* Linux/x86-64. */ - return tdesc_amd64_linux; + xcr0 = i386_linux_core_read_xcr0 (gdbarch, target, abfd); + if ((xcr0 & I386_XSTATE_AVX_MASK) == I386_XSTATE_AVX_MASK) + return tdesc_amd64_avx_linux; + else + return tdesc_amd64_linux; } static void @@ -1297,6 +1331,8 @@ amd64_linux_init_abi (struct gdbarch_info info, struct gdbarch *gdbarch) tdep->sc_reg_offset = amd64_linux_sc_reg_offset; tdep->sc_num_regs = ARRAY_SIZE (amd64_linux_sc_reg_offset); + tdep->xsave_xcr0_offset = I386_LINUX_XSAVE_XCR0_OFFSET; + /* GNU/Linux uses SVR4-style shared libraries. */ set_solib_svr4_fetch_link_map_offsets (gdbarch, svr4_lp64_fetch_link_map_offsets); @@ -1318,6 +1354,9 @@ amd64_linux_init_abi (struct gdbarch_info info, struct gdbarch *gdbarch) /* GNU/Linux uses SVR4-style shared libraries. */ set_gdbarch_skip_trampoline_code (gdbarch, find_solib_trampoline_target); + /* Install supported register note sections. */ + set_gdbarch_core_regset_sections (gdbarch, amd64_linux_regset_sections); + set_gdbarch_core_read_description (gdbarch, amd64_linux_core_read_description); @@ -1517,4 +1556,5 @@ _initialize_amd64_linux_tdep (void) /* Initialize the Linux target description */ initialize_tdesc_amd64_linux (); + initialize_tdesc_amd64_avx_linux (); } diff --git a/gdb/amd64-linux-tdep.h b/gdb/amd64-linux-tdep.h index 33316fb..8862057 100644 --- a/gdb/amd64-linux-tdep.h +++ b/gdb/amd64-linux-tdep.h @@ -26,13 +26,17 @@ /* Register number for the "orig_rax" register. If this register contains a value >= 0 it is interpreted as the system call number that the kernel is supposed to restart. */ -#define AMD64_LINUX_ORIG_RAX_REGNUM (AMD64_MXCSR_REGNUM + 1) +#define AMD64_LINUX_ORIG_RAX_REGNUM (AMD64_YMM15H_REGNUM + 1) /* Total number of registers for GNU/Linux. */ #define AMD64_LINUX_NUM_REGS (AMD64_LINUX_ORIG_RAX_REGNUM + 1) /* Linux target description. */ extern struct target_desc *tdesc_amd64_linux; +extern struct target_desc *tdesc_amd64_avx_linux; + +/* Update XSAVE extended state register note section. */ +extern void amd64_linux_update_xstateregset (unsigned int xstate_size); /* Enum that defines the syscall identifiers for amd64 linux. Used for process record/replay, these will be translated into diff --git a/gdb/amd64-tdep.c b/gdb/amd64-tdep.c index e5cfa71..aa4acfb 100644 --- a/gdb/amd64-tdep.c +++ b/gdb/amd64-tdep.c @@ -43,6 +43,7 @@ #include "i387-tdep.h" #include "features/i386/amd64.c" +#include "features/i386/amd64-avx.c" /* Note that the AMD64 architecture was previously known as x86-64. The latter is (forever) engraved into the canonical system name as @@ -71,8 +72,21 @@ static const char *amd64_register_names[] = "mxcsr", }; -/* Total number of registers. */ -#define AMD64_NUM_REGS ARRAY_SIZE (amd64_register_names) +static const char *amd64_ymm_names[] = +{ + "ymm0", "ymm1", "ymm2", "ymm3", + "ymm4", "ymm5", "ymm6", "ymm7", + "ymm8", "ymm9", "ymm10", "ymm11", + "ymm12", "ymm13", "ymm14", "ymm15" +}; + +static const char *amd64_ymmh_names[] = +{ + "ymm0h", "ymm1h", "ymm2h", "ymm3h", + "ymm4h", "ymm5h", "ymm6h", "ymm7h", + "ymm8h", "ymm9h", "ymm10h", "ymm11h", + "ymm12h", "ymm13h", "ymm14h", "ymm15h" +}; /* The registers used to pass integer arguments during a function call. */ static int amd64_dummy_call_integer_regs[] = @@ -163,6 +177,8 @@ static const int amd64_dwarf_regmap_len = static int amd64_dwarf_reg_to_regnum (struct gdbarch *gdbarch, int reg) { + struct gdbarch_tdep *tdep = gdbarch_tdep (gdbarch); + int ymm0_regnum = tdep->ymm0_regnum; int regnum = -1; if (reg >= 0 && reg < amd64_dwarf_regmap_len) @@ -170,6 +186,9 @@ amd64_dwarf_reg_to_regnum (struct gdbarch *gdbarch, int reg) if (regnum == -1) warning (_("Unmapped DWARF Register #%d encountered."), reg); + else if (ymm0_regnum >= 0 + && i386_xmm_regnum_p (gdbarch, regnum)) + regnum += ymm0_regnum - I387_XMM0_REGNUM (tdep); return regnum; } @@ -234,6 +253,19 @@ static const char *amd64_dword_names[] = "r8d", "r9d", "r10d", "r11d", "r12d", "r13d", "r14d", "r15d" }; +/* Return the name of register REGNUM, or the empty string if it is + an anonymous register. */ + +static const char * +amd64_register_name (struct gdbarch *gdbarch, int regnum) +{ + /* Hide the upper YMM registers. */ + if (i386_ymmh_regnum_p (gdbarch, regnum)) + return ""; + + return tdesc_register_name (gdbarch, regnum); +} + /* Return the name of register REGNUM. */ static const char * @@ -242,6 +274,8 @@ amd64_pseudo_register_name (struct gdbarch *gdbarch, int regnum) struct gdbarch_tdep *tdep = gdbarch_tdep (gdbarch); if (i386_byte_regnum_p (gdbarch, regnum)) return amd64_byte_names[regnum - tdep->al_regnum]; + else if (i386_ymm_regnum_p (gdbarch, regnum)) + return amd64_ymm_names[regnum - tdep->ymm0_regnum]; else if (i386_word_regnum_p (gdbarch, regnum)) return amd64_word_names[regnum - tdep->ax_regnum]; else if (i386_dword_regnum_p (gdbarch, regnum)) @@ -2148,6 +2182,28 @@ amd64_collect_fpregset (const struct regset *regset, amd64_collect_fxsave (regcache, regnum, fpregs); } +/* Similar to amd64_supply_fpregset, but use XSAVE extended state. */ + +static void +amd64_supply_xstateregset (const struct regset *regset, + struct regcache *regcache, int regnum, + const void *xstateregs, size_t len) +{ + const struct gdbarch_tdep *tdep = gdbarch_tdep (regset->arch); + amd64_supply_xsave (regcache, regnum, xstateregs); +} + +/* Similar to amd64_collect_fpregset, but use XSAVE extended state. */ + +static void +amd64_collect_xstateregset (const struct regset *regset, + const struct regcache *regcache, + int regnum, void *xstateregs, size_t len) +{ + const struct gdbarch_tdep *tdep = gdbarch_tdep (regset->arch); + amd64_collect_xsave (regcache, regnum, xstateregs, 1); +} + /* Return the appropriate register set for the core section identified by SECT_NAME and SECT_SIZE. */ @@ -2166,6 +2222,16 @@ amd64_regset_from_core_section (struct gdbarch *gdbarch, return tdep->fpregset; } + if (strcmp (sect_name, ".reg-xstate") == 0) + { + if (tdep->xstateregset == NULL) + tdep->xstateregset = regset_alloc (gdbarch, + amd64_supply_xstateregset, + amd64_collect_xstateregset); + + return tdep->xstateregset; + } + return i386_regset_from_core_section (gdbarch, sect_name, sect_size); } @@ -2228,6 +2294,13 @@ amd64_init_abi (struct gdbarch_info info, struct gdbarch *gdbarch) tdep->num_core_regs = AMD64_NUM_GREGS + I387_NUM_REGS; tdep->register_names = amd64_register_names; + if (tdesc_find_feature (tdesc, "org.gnu.gdb.i386.avx") != NULL) + { + tdep->ymmh_register_names = amd64_ymmh_names; + tdep->num_ymm_regs = 16; + tdep->ymm0h_regnum = AMD64_YMM0H_REGNUM; + } + tdep->num_byte_regs = 16; tdep->num_word_regs = 16; tdep->num_dword_regs = 16; @@ -2241,6 +2314,8 @@ amd64_init_abi (struct gdbarch_info info, struct gdbarch *gdbarch) set_tdesc_pseudo_register_name (gdbarch, amd64_pseudo_register_name); + set_gdbarch_register_name (gdbarch, amd64_register_name); + /* AMD64 has an FPU and 16 SSE registers. */ tdep->st0_regnum = AMD64_ST0_REGNUM; tdep->num_xmm_regs = 16; @@ -2321,6 +2396,7 @@ void _initialize_amd64_tdep (void) { initialize_tdesc_amd64 (); + initialize_tdesc_amd64_avx (); } @@ -2356,6 +2432,30 @@ amd64_supply_fxsave (struct regcache *regcache, int regnum, } } +/* Similar to amd64_supply_fxsave, but use XSAVE extended state. */ + +void +amd64_supply_xsave (struct regcache *regcache, int regnum, + const void *xsave) +{ + struct gdbarch *gdbarch = get_regcache_arch (regcache); + struct gdbarch_tdep *tdep = gdbarch_tdep (gdbarch); + + i387_supply_xsave (regcache, regnum, xsave); + + if (xsave && gdbarch_ptr_bit (gdbarch) == 64) + { + const gdb_byte *regs = xsave; + + if (regnum == -1 || regnum == I387_FISEG_REGNUM (tdep)) + regcache_raw_supply (regcache, I387_FISEG_REGNUM (tdep), + regs + 12); + if (regnum == -1 || regnum == I387_FOSEG_REGNUM (tdep)) + regcache_raw_supply (regcache, I387_FOSEG_REGNUM (tdep), + regs + 20); + } +} + /* Fill register REGNUM (if it is a floating-point or SSE register) in *FXSAVE with the value from REGCACHE. If REGNUM is -1, do this for all registers. This function doesn't touch any of the reserved @@ -2379,3 +2479,26 @@ amd64_collect_fxsave (const struct regcache *regcache, int regnum, regcache_raw_collect (regcache, I387_FOSEG_REGNUM (tdep), regs + 20); } } + +/* Similar to amd64_collect_fxsave, but but use XSAVE extended state. */ + +void +amd64_collect_xsave (const struct regcache *regcache, int regnum, + void *xsave, int gcore) +{ + struct gdbarch *gdbarch = get_regcache_arch (regcache); + struct gdbarch_tdep *tdep = gdbarch_tdep (gdbarch); + gdb_byte *regs = xsave; + + i387_collect_xsave (regcache, regnum, xsave, gcore); + + if (gdbarch_ptr_bit (gdbarch) == 64) + { + if (regnum == -1 || regnum == I387_FISEG_REGNUM (tdep)) + regcache_raw_collect (regcache, I387_FISEG_REGNUM (tdep), + regs + 12); + if (regnum == -1 || regnum == I387_FOSEG_REGNUM (tdep)) + regcache_raw_collect (regcache, I387_FOSEG_REGNUM (tdep), + regs + 20); + } +} diff --git a/gdb/amd64-tdep.h b/gdb/amd64-tdep.h index 363479c..9f07dda 100644 --- a/gdb/amd64-tdep.h +++ b/gdb/amd64-tdep.h @@ -61,12 +61,16 @@ enum amd64_regnum AMD64_FSTAT_REGNUM = AMD64_ST0_REGNUM + 9, AMD64_XMM0_REGNUM = 40, /* %xmm0 */ AMD64_XMM1_REGNUM, /* %xmm1 */ - AMD64_MXCSR_REGNUM = AMD64_XMM0_REGNUM + 16 + AMD64_MXCSR_REGNUM = AMD64_XMM0_REGNUM + 16, + AMD64_YMM0H_REGNUM, /* %ymm0h */ + AMD64_YMM15H_REGNUM = AMD64_YMM0H_REGNUM + 15 }; /* Number of general purpose registers. */ #define AMD64_NUM_GREGS 24 +#define AMD64_NUM_REGS (AMD64_YMM15H_REGNUM + 1) + extern struct displaced_step_closure *amd64_displaced_step_copy_insn (struct gdbarch *gdbarch, CORE_ADDR from, CORE_ADDR to, struct regcache *regs); @@ -77,12 +81,6 @@ extern void amd64_displaced_step_fixup (struct gdbarch *gdbarch, extern void amd64_init_abi (struct gdbarch_info info, struct gdbarch *gdbarch); -/* Functions from amd64-tdep.c which may be needed on architectures - with extra registers. */ - -extern const char *amd64_register_name (struct gdbarch *gdbarch, int regnum); -extern struct type *amd64_register_type (struct gdbarch *gdbarch, int regnum); - /* Fill register REGNUM in REGCACHE with the appropriate floating-point or SSE register value from *FXSAVE. If REGNUM is -1, do this for all registers. This function masks off any of the @@ -91,6 +89,10 @@ extern struct type *amd64_register_type (struct gdbarch *gdbarch, int regnum); extern void amd64_supply_fxsave (struct regcache *regcache, int regnum, const void *fxsave); +/* Similar to amd64_supply_fxsave, but use XSAVE extended state. */ +extern void amd64_supply_xsave (struct regcache *regcache, int regnum, + const void *xsave); + /* Fill register REGNUM (if it is a floating-point or SSE register) in *FXSAVE with the value from REGCACHE. If REGNUM is -1, do this for all registers. This function doesn't touch any of the reserved @@ -99,6 +101,10 @@ extern void amd64_supply_fxsave (struct regcache *regcache, int regnum, extern void amd64_collect_fxsave (const struct regcache *regcache, int regnum, void *fxsave); +/* Similar to amd64_collect_fxsave, but but use XSAVE extended state. */ +extern void amd64_collect_xsave (const struct regcache *regcache, + int regnum, void *xsave, int gcore); + void amd64_classify (struct type *type, enum amd64_reg_class class[2]); diff --git a/gdb/testsuite/gdb.arch/i386-avx.c b/gdb/testsuite/gdb.arch/i386-avx.c new file mode 100644 index 0000000..73f92b6 --- /dev/null +++ b/gdb/testsuite/gdb.arch/i386-avx.c @@ -0,0 +1,128 @@ +/* Test program for AVX registers. + + Copyright 2010 Free Software Foundation, Inc. + + This file is part of GDB. + + 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 3 of the License, or + (at your option) any later version. + + 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, see . */ + +#include +#include "i386-cpuid.h" + +typedef struct { + float f[8]; +} v8sf_t; + + +v8sf_t data[] = + { + { { 0.0, 0.125, 0.25, 0.375, 0.50, 0.625, 0.75, 0.875 } }, + { { 1.0, 1.125, 1.25, 1.375, 1.50, 1.625, 1.75, 1.875 } }, + { { 2.0, 2.125, 2.25, 2.375, 2.50, 2.625, 2.75, 2.875 } }, + { { 3.0, 3.125, 3.25, 3.375, 3.50, 3.625, 3.75, 3.875 } }, + { { 4.0, 4.125, 4.25, 4.375, 4.50, 4.625, 4.75, 4.875 } }, + { { 5.0, 5.125, 5.25, 5.375, 5.50, 5.625, 5.75, 5.875 } }, + { { 6.0, 6.125, 6.25, 6.375, 6.50, 6.625, 6.75, 6.875 } }, + { { 7.0, 7.125, 7.25, 7.375, 7.50, 7.625, 7.75, 7.875 } }, +#ifdef __x86_64__ + { { 8.0, 8.125, 8.25, 8.375, 8.50, 8.625, 8.75, 8.875 } }, + { { 9.0, 9.125, 9.25, 9.375, 9.50, 9.625, 9.75, 9.875 } }, + { { 10.0, 10.125, 10.25, 10.375, 10.50, 10.625, 10.75, 10.875 } }, + { { 11.0, 11.125, 11.25, 11.375, 11.50, 11.625, 11.75, 11.875 } }, + { { 12.0, 12.125, 12.25, 12.375, 12.50, 12.625, 12.75, 12.875 } }, + { { 13.0, 13.125, 13.25, 13.375, 13.50, 13.625, 13.75, 13.875 } }, + { { 14.0, 14.125, 14.25, 14.375, 14.50, 14.625, 14.75, 14.875 } }, + { { 15.0, 15.125, 15.25, 15.375, 15.50, 15.625, 15.75, 15.875 } }, +#endif + }; + + +int +have_avx (void) +{ + unsigned int eax, ebx, ecx, edx; + + if (!__get_cpuid (1, &eax, &ebx, &ecx, &edx)) + return 0; + + if ((ecx & (bit_AVX | bit_OSXSAVE)) == (bit_AVX | bit_OSXSAVE)) + return 1; + else + return 0; +} + +int +main (int argc, char **argv) +{ + if (have_avx ()) + { + asm ("vmovaps 0(%0), %%ymm0\n\t" + "vmovaps 32(%0), %%ymm1\n\t" + "vmovaps 64(%0), %%ymm2\n\t" + "vmovaps 96(%0), %%ymm3\n\t" + "vmovaps 128(%0), %%ymm4\n\t" + "vmovaps 160(%0), %%ymm5\n\t" + "vmovaps 192(%0), %%ymm6\n\t" + "vmovaps 224(%0), %%ymm7\n\t" + : /* no output operands */ + : "r" (data) + : "xmm0", "xmm1", "xmm2", "xmm3", "xmm4", "xmm5", "xmm6", "xmm7"); +#ifdef __x86_64__ + asm ("vmovaps 256(%0), %%ymm8\n\t" + "vmovaps 288(%0), %%ymm9\n\t" + "vmovaps 320(%0), %%ymm10\n\t" + "vmovaps 352(%0), %%ymm11\n\t" + "vmovaps 384(%0), %%ymm12\n\t" + "vmovaps 416(%0), %%ymm13\n\t" + "vmovaps 448(%0), %%ymm14\n\t" + "vmovaps 480(%0), %%ymm15\n\t" + : /* no output operands */ + : "r" (data) + : "xmm8", "xmm9", "xmm10", "xmm11", "xmm12", "xmm13", "xmm14", "xmm15"); +#endif + + asm ("nop"); /* first breakpoint here */ + + asm ( + "vmovaps %%ymm0, 0(%0)\n\t" + "vmovaps %%ymm1, 32(%0)\n\t" + "vmovaps %%ymm2, 64(%0)\n\t" + "vmovaps %%ymm3, 96(%0)\n\t" + "vmovaps %%ymm4, 128(%0)\n\t" + "vmovaps %%ymm5, 160(%0)\n\t" + "vmovaps %%ymm6, 192(%0)\n\t" + "vmovaps %%ymm7, 224(%0)\n\t" + : /* no output operands */ + : "r" (data) + : "xmm0", "xmm1", "xmm2", "xmm3", "xmm4", "xmm5", "xmm6", "xmm7"); +#ifdef __x86_64__ + asm ( + "vmovaps %%ymm8, 256(%0)\n\t" + "vmovaps %%ymm9, 288(%0)\n\t" + "vmovaps %%ymm10, 320(%0)\n\t" + "vmovaps %%ymm11, 352(%0)\n\t" + "vmovaps %%ymm12, 384(%0)\n\t" + "vmovaps %%ymm13, 416(%0)\n\t" + "vmovaps %%ymm14, 448(%0)\n\t" + "vmovaps %%ymm15, 480(%0)\n\t" + : /* no output operands */ + : "r" (data) + : "xmm8", "xmm9", "xmm10", "xmm11", "xmm12", "xmm13", "xmm14", "xmm15"); +#endif + + puts ("Bye!"); /* second breakpoint here */ + } + + return 0; +} diff --git a/gdb/testsuite/gdb.arch/i386-avx.exp b/gdb/testsuite/gdb.arch/i386-avx.exp new file mode 100644 index 0000000..561ddef --- /dev/null +++ b/gdb/testsuite/gdb.arch/i386-avx.exp @@ -0,0 +1,110 @@ +# Copyright 2010 Free Software Foundation, Inc. + +# 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 3 of the License, or +# (at your option) any later version. +# +# 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, see . + +# Please email any bugs, comments, and/or additions to this file to: +# bug-gdb@gnu.org + +# This file is part of the gdb testsuite. + +if $tracelevel { + strace $tracelevel +} + +set prms_id 0 +set bug_id 0 + +if { ![istarget i?86-*-*] && ![istarget x86_64-*-* ] } { + verbose "Skipping x86 AVX tests." + return +} + +set testfile "i386-avx" +set srcfile ${testfile}.c +set binfile ${objdir}/${subdir}/${testfile} + +if [get_compiler_info ${binfile}] { + return -1 +} + +set additional_flags "" +if [test_compiler_info gcc*] { + set additional_flags "additional_flags=-mavx" +} + +if { [gdb_compile "${srcdir}/${subdir}/${srcfile}" "${binfile}" executable [list debug $additional_flags]] != "" } { + unsupported "compiler does not support AVX" + return +} + +gdb_exit +gdb_start +gdb_reinitialize_dir $srcdir/$subdir +gdb_load ${binfile} + +if ![runto_main] then { + gdb_suppress_tests +} + +send_gdb "print have_avx ()\r" +gdb_expect { + -re ".. = 1\r\n$gdb_prompt " { + pass "check whether processor supports AVX" + } + -re ".. = 0\r\n$gdb_prompt " { + verbose "processor does not support AVX; skipping AVX tests" + return + } + -re ".*$gdb_prompt $" { + fail "check whether processor supports AVX" + } + timeout { + fail "check whether processor supports AVX (timeout)" + } +} + +gdb_test "break [gdb_get_line_number "first breakpoint here"]" \ + "Breakpoint .* at .*i386-avx.c.*" \ + "set first breakpoint in main" +gdb_continue_to_breakpoint "continue to first breakpoint in main" + +if [istarget i?86-*-*] { + set nr_regs 8 +} else { + set nr_regs 16 +} + +for { set r 0 } { $r < $nr_regs } { incr r } { + gdb_test "print \$ymm$r.v8_float" \ + ".. = \\{$r, $r.125, $r.25, $r.375, $r.5, $r.625, $r.75, $r.875\\}.*" \ + "check float contents of %ymm$r" + gdb_test "print \$ymm$r.v32_int8" \ + ".. = \\{(-?\[0-9\]+, ){31}-?\[0-9\]+\\}.*" \ + "check int8 contents of %ymm$r" +} + +for { set r 0 } { $r < $nr_regs } { incr r } { + gdb_test "set var \$ymm$r.v8_float\[0\] = $r + 10" "" "set %ymm$r" +} + +gdb_test "break [gdb_get_line_number "second breakpoint here"]" \ + "Breakpoint .* at .*i386-avx.c.*" \ + "set second breakpoint in main" +gdb_continue_to_breakpoint "continue to second breakpoint in main" + +for { set r 0 } { $r < $nr_regs } { incr r } { + gdb_test "print data\[$r\]" \ + ".. = \\{f = \\{[expr $r + 10], $r.125, $r.25, $r.375, $r.5, $r.625, $r.75, $r.875\\}\\}.*" \ + "check contents of data\[$r\]" +} diff --git a/gdb/testsuite/gdb.arch/i386-cpuid.h b/gdb/testsuite/gdb.arch/i386-cpuid.h index 7ff0dba..5ebde5a 100644 --- a/gdb/testsuite/gdb.arch/i386-cpuid.h +++ b/gdb/testsuite/gdb.arch/i386-cpuid.h @@ -1,75 +1,200 @@ -/* Helper file for i386 platform. Runtime check for MMX/SSE/SSE2 support. +/* Helper file for i386 platform. Runtime check for MMX/SSE/SSE2/AVX + * support. Copied from gcc 4.4. + * + * Copyright (C) 2007, 2008, 2009 Free Software Foundation, Inc. + * + * This file 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 3, or (at your option) any + * later version. + * + * This file 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. + * + * Under Section 7 of GPL version 3, you are granted additional + * permissions described in the GCC Runtime Library Exception, version + * 3.1, as published by the Free Software Foundation. + * + * You should have received a copy of the GNU General Public License and + * a copy of the GCC Runtime Library Exception along with this program; + * see the files COPYING3 and COPYING.RUNTIME respectively. If not, see + * . + */ - Copyright 2004, 2007, 2008, 2009, 2010 Free Software Foundation, Inc. +/* %ecx */ +#define bit_SSE3 (1 << 0) +#define bit_PCLMUL (1 << 1) +#define bit_SSSE3 (1 << 9) +#define bit_FMA (1 << 12) +#define bit_CMPXCHG16B (1 << 13) +#define bit_SSE4_1 (1 << 19) +#define bit_SSE4_2 (1 << 20) +#define bit_MOVBE (1 << 22) +#define bit_POPCNT (1 << 23) +#define bit_AES (1 << 25) +#define bit_XSAVE (1 << 26) +#define bit_OSXSAVE (1 << 27) +#define bit_AVX (1 << 28) - This file is part of GDB. +/* %edx */ +#define bit_CMPXCHG8B (1 << 8) +#define bit_CMOV (1 << 15) +#define bit_MMX (1 << 23) +#define bit_FXSAVE (1 << 24) +#define bit_SSE (1 << 25) +#define bit_SSE2 (1 << 26) - 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 3 of the License, or - (at your option) any later version. +/* Extended Features */ +/* %ecx */ +#define bit_LAHF_LM (1 << 0) +#define bit_ABM (1 << 5) +#define bit_SSE4a (1 << 6) +#define bit_XOP (1 << 11) +#define bit_LWP (1 << 15) +#define bit_FMA4 (1 << 16) - 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. +/* %edx */ +#define bit_LM (1 << 29) +#define bit_3DNOWP (1 << 30) +#define bit_3DNOW (1 << 31) - You should have received a copy of the GNU General Public License - along with this program. If not, see . */ -/* Used by 20020523-2.c and i386-sse-6.c, and possibly others. */ -/* Plagarized from 20020523-2.c. */ -/* Plagarized from gcc. */ +#if defined(__i386__) && defined(__PIC__) +/* %ebx may be the PIC register. */ +#if __GNUC__ >= 3 +#define __cpuid(level, a, b, c, d) \ + __asm__ ("xchg{l}\t{%%}ebx, %1\n\t" \ + "cpuid\n\t" \ + "xchg{l}\t{%%}ebx, %1\n\t" \ + : "=a" (a), "=r" (b), "=c" (c), "=d" (d) \ + : "0" (level)) -#define bit_CMOV (1 << 15) -#define bit_MMX (1 << 23) -#define bit_SSE (1 << 25) -#define bit_SSE2 (1 << 26) +#define __cpuid_count(level, count, a, b, c, d) \ + __asm__ ("xchg{l}\t{%%}ebx, %1\n\t" \ + "cpuid\n\t" \ + "xchg{l}\t{%%}ebx, %1\n\t" \ + : "=a" (a), "=r" (b), "=c" (c), "=d" (d) \ + : "0" (level), "2" (count)) +#else +/* Host GCCs older than 3.0 weren't supporting Intel asm syntax + nor alternatives in i386 code. */ +#define __cpuid(level, a, b, c, d) \ + __asm__ ("xchgl\t%%ebx, %1\n\t" \ + "cpuid\n\t" \ + "xchgl\t%%ebx, %1\n\t" \ + : "=a" (a), "=r" (b), "=c" (c), "=d" (d) \ + : "0" (level)) -#ifndef NOINLINE -#define NOINLINE __attribute__ ((noinline)) +#define __cpuid_count(level, count, a, b, c, d) \ + __asm__ ("xchgl\t%%ebx, %1\n\t" \ + "cpuid\n\t" \ + "xchgl\t%%ebx, %1\n\t" \ + : "=a" (a), "=r" (b), "=c" (c), "=d" (d) \ + : "0" (level), "2" (count)) #endif +#else +#define __cpuid(level, a, b, c, d) \ + __asm__ ("cpuid\n\t" \ + : "=a" (a), "=b" (b), "=c" (c), "=d" (d) \ + : "0" (level)) -unsigned int i386_cpuid (void) NOINLINE; +#define __cpuid_count(level, count, a, b, c, d) \ + __asm__ ("cpuid\n\t" \ + : "=a" (a), "=b" (b), "=c" (c), "=d" (d) \ + : "0" (level), "2" (count)) +#endif -unsigned int NOINLINE -i386_cpuid (void) +/* Return highest supported input value for cpuid instruction. ext can + be either 0x0 or 0x8000000 to return highest supported value for + basic or extended cpuid information. Function returns 0 if cpuid + is not supported or whatever cpuid returns in eax register. If sig + pointer is non-null, then first four bytes of the signature + (as found in ebx register) are returned in location pointed by sig. */ + +static __inline unsigned int +__get_cpuid_max (unsigned int __ext, unsigned int *__sig) { - int fl1, fl2; + unsigned int __eax, __ebx, __ecx, __edx; #ifndef __x86_64__ +#if __GNUC__ >= 3 /* See if we can use cpuid. On AMD64 we always can. */ - __asm__ ("pushfl; pushfl; popl %0; movl %0,%1; xorl %2,%0;" - "pushl %0; popfl; pushfl; popl %0; popfl" - : "=&r" (fl1), "=&r" (fl2) + __asm__ ("pushf{l|d}\n\t" + "pushf{l|d}\n\t" + "pop{l}\t%0\n\t" + "mov{l}\t{%0, %1|%1, %0}\n\t" + "xor{l}\t{%2, %0|%0, %2}\n\t" + "push{l}\t%0\n\t" + "popf{l|d}\n\t" + "pushf{l|d}\n\t" + "pop{l}\t%0\n\t" + "popf{l|d}\n\t" + : "=&r" (__eax), "=&r" (__ebx) + : "i" (0x00200000)); +#else +/* Host GCCs older than 3.0 weren't supporting Intel asm syntax + nor alternatives in i386 code. */ + __asm__ ("pushfl\n\t" + "pushfl\n\t" + "popl\t%0\n\t" + "movl\t%0, %1\n\t" + "xorl\t%2, %0\n\t" + "pushl\t%0\n\t" + "popfl\n\t" + "pushfl\n\t" + "popl\t%0\n\t" + "popfl\n\t" + : "=&r" (__eax), "=&r" (__ebx) : "i" (0x00200000)); - if (((fl1 ^ fl2) & 0x00200000) == 0) - return (0); #endif - /* Host supports cpuid. See if cpuid gives capabilities, try - CPUID(0). Preserve %ebx and %ecx; cpuid insn clobbers these, we - don't need their CPUID values here, and %ebx may be the PIC - register. */ -#ifdef __x86_64__ - __asm__ ("pushq %%rcx; pushq %%rbx; cpuid; popq %%rbx; popq %%rcx" - : "=a" (fl1) : "0" (0) : "rdx", "cc"); -#else - __asm__ ("pushl %%ecx; pushl %%ebx; cpuid; popl %%ebx; popl %%ecx" - : "=a" (fl1) : "0" (0) : "edx", "cc"); + if (!((__eax ^ __ebx) & 0x00200000)) + return 0; #endif - if (fl1 == 0) - return (0); - - /* Invoke CPUID(1), return %edx; caller can examine bits to - determine what's supported. */ -#ifdef __x86_64__ - __asm__ ("pushq %%rcx; pushq %%rbx; cpuid; popq %%rbx; popq %%rcx" - : "=d" (fl2), "=a" (fl1) : "1" (1) : "cc"); -#else - __asm__ ("pushl %%ecx; pushl %%ebx; cpuid; popl %%ebx; popl %%ecx" - : "=d" (fl2), "=a" (fl1) : "1" (1) : "cc"); + + /* Host supports cpuid. Return highest supported cpuid input value. */ + __cpuid (__ext, __eax, __ebx, __ecx, __edx); + + if (__sig) + *__sig = __ebx; + + return __eax; +} + +/* Return cpuid data for requested cpuid level, as found in returned + eax, ebx, ecx and edx registers. The function checks if cpuid is + supported and returns 1 for valid cpuid information or 0 for + unsupported cpuid level. All pointers are required to be non-null. */ + +static __inline int +__get_cpuid (unsigned int __level, + unsigned int *__eax, unsigned int *__ebx, + unsigned int *__ecx, unsigned int *__edx) +{ + unsigned int __ext = __level & 0x80000000; + + if (__get_cpuid_max (__ext, 0) < __level) + return 0; + + __cpuid (__level, *__eax, *__ebx, *__ecx, *__edx); + return 1; +} + +#ifndef NOINLINE +#define NOINLINE __attribute__ ((noinline)) #endif - return fl2; +unsigned int i386_cpuid (void) NOINLINE; + +unsigned int NOINLINE +i386_cpuid (void) +{ + unsigned int eax, ebx, ecx, edx; + + if (!__get_cpuid (1, &eax, &ebx, &ecx, &edx)) + return 0; + + return edx; }