mirror of
https://github.com/jart/cosmopolitan.git
synced 2024-05-18 03:22:40 +00:00
957c61cbbf
This change upgrades to GCC 12.3 and GNU binutils 2.42. The GNU linker appears to have changed things so that only a single de-duplicated str table is present in the binary, and it gets placed wherever the linker wants, regardless of what the linker script says. To cope with that we need to stop using .ident to embed licenses. As such, this change does significant work to revamp how third party licenses are defined in the codebase, using `.section .notice,"aR",@progbits`. This new GCC 12.3 toolchain has support for GNU indirect functions. It lets us support __target_clones__ for the first time. This is used for optimizing the performance of libc string functions such as strlen and friends so far on x86, by ensuring AVX systems favor a second codepath that uses VEX encoding. It shaves some latency off certain operations. It's a useful feature to have for scientific computing for the reasons explained by the test/libcxx/openmp_test.cc example which compiles for fifteen different microarchitectures. Thanks to the upgrades, it's now also possible to use newer instruction sets, such as AVX512FP16, VNNI. Cosmo now uses the %gs register on x86 by default for TLS. Doing it is helpful for any program that links `cosmo_dlopen()`. Such programs had to recompile their binaries at startup to change the TLS instructions. That's not great, since it means every page in the executable needs to be faulted. The work of rewriting TLS-related x86 opcodes, is moved to fixupobj.com instead. This is great news for MacOS x86 users, since we previously needed to morph the binary every time for that platform but now that's no longer necessary. The only platforms where we need fixup of TLS x86 opcodes at runtime are now Windows, OpenBSD, and NetBSD. On Windows we morph TLS to point deeper into the TIB, based on a TlsAlloc assignment, and on OpenBSD/NetBSD we morph %gs back into %fs since the kernels do not allow us to specify a value for the %gs register. OpenBSD users are now required to use APE Loader to run Cosmo binaries and assimilation is no longer possible. OpenBSD kernel needs to change to allow programs to specify a value for the %gs register, or it needs to stop marking executable pages loaded by the kernel as mimmutable(). This release fixes __constructor__, .ctor, .init_array, and lastly the .preinit_array so they behave the exact same way as glibc. We no longer use hex constants to define math.h symbols like M_PI.
90 lines
2.5 KiB
C
90 lines
2.5 KiB
C
#ifndef COSMOPOLITAN_LIBC_THREAD_TLS_H_
|
|
#define COSMOPOLITAN_LIBC_THREAD_TLS_H_
|
|
|
|
#define TLS_ALIGNMENT 64
|
|
|
|
#define TIB_FLAG_VFORKED 1
|
|
|
|
#if !(__ASSEMBLER__ + __LINKER__ + 0)
|
|
COSMOPOLITAN_C_START_
|
|
|
|
struct CosmoFtrace { /* 16 */
|
|
char ft_once; /* 0 */
|
|
char ft_noreentry; /* 1 */
|
|
int ft_skew; /* 4 */
|
|
int64_t ft_lastaddr; /* 8 */
|
|
};
|
|
|
|
/* NOTE: update aarch64 libc/errno.h if sizeof changes */
|
|
/* NOTE: update aarch64 libc/proc/vfork.S if sizeof changes */
|
|
/* NOTE: update aarch64 libc/nexgen32e/gc.S if sizeof changes */
|
|
struct CosmoTib {
|
|
struct CosmoTib *tib_self; /* 0x00 */
|
|
struct CosmoFtrace tib_ftracer; /* 0x08 */
|
|
void *tib_garbages; /* 0x18 */
|
|
intptr_t tib_locale; /* 0x20 */
|
|
intptr_t tib_pthread; /* 0x28 */
|
|
struct CosmoTib *tib_self2; /* 0x30 */
|
|
_Atomic(int32_t) tib_tid; /* 0x38 transitions -1 → tid → 0 */
|
|
int32_t tib_errno; /* 0x3c */
|
|
uint64_t tib_flags; /* 0x40 */
|
|
int tib_ftrace; /* inherited */
|
|
int tib_strace; /* inherited */
|
|
_Atomic(uint64_t) tib_sigmask; /* inherited */
|
|
_Atomic(uint64_t) tib_sigpending;
|
|
_Atomic(uint64_t) tib_syshand; /* win32=kThread, xnusilicon=pthread_t */
|
|
char *tib_sigstack_addr;
|
|
uint32_t tib_sigstack_size;
|
|
uint32_t tib_sigstack_flags;
|
|
void **tib_keys;
|
|
void *tib_nsync;
|
|
void *tib_todo[7];
|
|
} __attribute__((__aligned__(64)));
|
|
|
|
extern int __threaded;
|
|
extern char __tls_morphed;
|
|
extern unsigned __tls_index;
|
|
|
|
char *_mktls(struct CosmoTib **) libcesque;
|
|
void __bootstrap_tls(struct CosmoTib *, char *) libcesque;
|
|
|
|
#ifdef __x86_64__
|
|
extern char __tls_enabled;
|
|
#define __tls_enabled_set(x) __tls_enabled = x
|
|
#elif defined(__aarch64__)
|
|
#define __tls_enabled true
|
|
#define __tls_enabled_set(x) (void)0
|
|
#else
|
|
#error "unsupported architecture"
|
|
#endif
|
|
|
|
void __set_tls(struct CosmoTib *) libcesque;
|
|
|
|
/**
|
|
* Returns location of thread information block.
|
|
*
|
|
* This can't be used in privileged functions.
|
|
*/
|
|
forceinline pureconst struct CosmoTib *__get_tls(void) {
|
|
#ifdef __chibicc__
|
|
return 0;
|
|
#elif __x86_64__
|
|
struct CosmoTib *__tib;
|
|
__asm__("mov\t%%gs:0x30,%0" : "=r"(__tib));
|
|
return __tib;
|
|
#elif defined(__aarch64__)
|
|
register struct CosmoTib *__tls __asm__("x28");
|
|
return __tls - 1;
|
|
#endif
|
|
}
|
|
|
|
#ifdef __x86_64__
|
|
#define __adj_tls(tib) (tib)
|
|
#elif defined(__aarch64__)
|
|
#define __adj_tls(tib) ((struct CosmoTib *)(tib) + 1)
|
|
#endif
|
|
|
|
COSMOPOLITAN_C_END_
|
|
#endif /* !(__ASSEMBLER__ + __LINKER__ + 0) */
|
|
#endif /* COSMOPOLITAN_LIBC_THREAD_TLS_H_ */
|