2020-06-15 14:18:57 +00:00
|
|
|
/*-*- mode:c;indent-tabs-mode:nil;c-basic-offset:2;tab-width:8;coding:utf-8 -*-│
|
|
|
|
│ vi: set et ft=c ts=2 sts=2 sw=2 fenc=utf-8 :vi │
|
|
|
|
╞══════════════════════════════════════════════════════════════════════════════╡
|
|
|
|
│ Copyright 2020 Justine Alexandra Roberts Tunney │
|
|
|
|
│ │
|
2020-12-28 01:18:44 +00:00
|
|
|
│ Permission to use, copy, modify, and/or distribute this software for │
|
|
|
|
│ any purpose with or without fee is hereby granted, provided that the │
|
|
|
|
│ above copyright notice and this permission notice appear in all copies. │
|
2020-06-15 14:18:57 +00:00
|
|
|
│ │
|
2020-12-28 01:18:44 +00:00
|
|
|
│ THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL │
|
|
|
|
│ WARRANTIES WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED │
|
|
|
|
│ WARRANTIES OF MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE │
|
|
|
|
│ AUTHOR BE LIABLE FOR ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL │
|
|
|
|
│ DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR │
|
|
|
|
│ PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER │
|
|
|
|
│ TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR │
|
|
|
|
│ PERFORMANCE OF THIS SOFTWARE. │
|
2020-06-15 14:18:57 +00:00
|
|
|
╚─────────────────────────────────────────────────────────────────────────────*/
|
|
|
|
#include "libc/calls/calls.h"
|
2022-11-06 02:49:41 +00:00
|
|
|
#include "libc/calls/cp.internal.h"
|
2020-06-15 14:18:57 +00:00
|
|
|
#include "libc/calls/internal.h"
|
|
|
|
#include "libc/calls/struct/iovec.h"
|
2022-08-13 20:11:56 +00:00
|
|
|
#include "libc/calls/struct/iovec.internal.h"
|
2022-05-23 22:06:11 +00:00
|
|
|
#include "libc/calls/syscall-sysv.internal.h"
|
2022-08-13 20:11:56 +00:00
|
|
|
#include "libc/errno.h"
|
2021-05-14 12:36:58 +00:00
|
|
|
#include "libc/intrin/asan.internal.h"
|
2022-05-12 13:43:59 +00:00
|
|
|
#include "libc/intrin/describeflags.internal.h"
|
2022-08-13 20:11:56 +00:00
|
|
|
#include "libc/intrin/likely.h"
|
2022-09-13 06:10:38 +00:00
|
|
|
#include "libc/intrin/strace.internal.h"
|
2022-08-13 20:11:56 +00:00
|
|
|
#include "libc/intrin/weaken.h"
|
2024-05-26 23:53:13 +00:00
|
|
|
#include "libc/limits.h"
|
|
|
|
#include "libc/mem/alloca.h"
|
|
|
|
#include "libc/runtime/stack.h"
|
2023-08-17 03:11:19 +00:00
|
|
|
#include "libc/runtime/zipos.internal.h"
|
2020-06-15 14:18:57 +00:00
|
|
|
#include "libc/sock/internal.h"
|
2024-05-26 23:53:13 +00:00
|
|
|
#include "libc/stdckdint.h"
|
2020-06-15 14:18:57 +00:00
|
|
|
#include "libc/sysv/errfuns.h"
|
|
|
|
|
2024-05-26 23:53:13 +00:00
|
|
|
static size_t SumIovecBytes(const struct iovec *iov, int iovlen) {
|
|
|
|
size_t count = 0;
|
|
|
|
for (int i = 0; i < iovlen; ++i)
|
|
|
|
if (ckd_add(&count, count, iov[i].iov_len))
|
|
|
|
count = SIZE_MAX;
|
|
|
|
return count;
|
|
|
|
}
|
|
|
|
|
|
|
|
static ssize_t readv_impl(int fd, const struct iovec *iov, int iovlen) {
|
|
|
|
if (fd < 0)
|
|
|
|
return ebadf();
|
|
|
|
if (iovlen < 0)
|
|
|
|
return einval();
|
|
|
|
if (IsAsan() && !__asan_is_valid_iov(iov, iovlen))
|
|
|
|
return efault();
|
|
|
|
|
|
|
|
// XNU and BSDs will EINVAL if requested bytes exceeds INT_MAX
|
|
|
|
// this is inconsistent with Linux which ignores huge requests
|
|
|
|
if (!IsLinux()) {
|
|
|
|
size_t sum, remain = 0x7ffff000;
|
|
|
|
if ((sum = SumIovecBytes(iov, iovlen)) > remain) {
|
|
|
|
struct iovec *iov2;
|
|
|
|
#pragma GCC push_options
|
|
|
|
#pragma GCC diagnostic ignored "-Walloca-larger-than="
|
|
|
|
iov2 = alloca(iovlen * sizeof(struct iovec));
|
|
|
|
CheckLargeStackAllocation(iov2, iovlen * sizeof(struct iovec));
|
|
|
|
#pragma GCC pop_options
|
|
|
|
for (int i = 0; i < iovlen; ++i) {
|
|
|
|
iov2[i] = iov[i];
|
|
|
|
if (remain >= iov2[i].iov_len) {
|
|
|
|
remain -= iov2[i].iov_len;
|
|
|
|
} else {
|
|
|
|
iov2[i].iov_len = remain;
|
|
|
|
remain = 0;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
iov = iov2;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
if (fd < g_fds.n && g_fds.p[fd].kind == kFdZip) {
|
|
|
|
return _weaken(__zipos_read)(
|
|
|
|
(struct ZiposHandle *)(intptr_t)g_fds.p[fd].handle, iov, iovlen, -1);
|
|
|
|
} else if (IsLinux() || IsXnu() || IsFreebsd() || IsOpenbsd() || IsNetbsd()) {
|
|
|
|
if (iovlen == 1) {
|
|
|
|
return sys_read(fd, iov[0].iov_base, iov[0].iov_len);
|
|
|
|
} else {
|
|
|
|
return sys_readv(fd, iov, iovlen);
|
|
|
|
}
|
|
|
|
} else if (fd >= g_fds.n) {
|
|
|
|
return ebadf();
|
|
|
|
} else if (IsMetal()) {
|
|
|
|
return sys_readv_metal(fd, iov, iovlen);
|
|
|
|
} else if (IsWindows()) {
|
|
|
|
return sys_readv_nt(fd, iov, iovlen);
|
|
|
|
} else {
|
|
|
|
return enosys();
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2020-06-15 14:18:57 +00:00
|
|
|
/**
|
|
|
|
* Reads data to multiple buffers.
|
|
|
|
*
|
2022-04-16 17:40:23 +00:00
|
|
|
* This is the same thing as read() except it has multiple buffers.
|
|
|
|
* This yields a performance boost in situations where it'd be expensive
|
|
|
|
* to stitch data together using memcpy() or issuing multiple syscalls.
|
|
|
|
* This wrapper is implemented so that readv() calls where iovlen<2 may
|
|
|
|
* be passed to the kernel as read() instead. This yields a 100 cycle
|
|
|
|
* performance boost in the case of a single small iovec.
|
|
|
|
*
|
2024-05-26 23:53:13 +00:00
|
|
|
* It's possible for file write request to be partially completed. For
|
|
|
|
* example, if the sum of `iov` lengths exceeds 0x7ffff000 then bytes
|
|
|
|
* beyond that will be ignored. This is a Linux behavior that Cosmo
|
|
|
|
* polyfills across platforms.
|
|
|
|
*
|
2020-06-15 14:18:57 +00:00
|
|
|
* @return number of bytes actually read, or -1 w/ errno
|
Make improvements
- We now serialize the file descriptor table when spawning / executing
processes on Windows. This means you can now inherit more stuff than
just standard i/o. It's needed by bash, which duplicates the console
to file descriptor #255. We also now do a better job serializing the
environment variables, so you're less likely to encounter E2BIG when
using your bash shell. We also no longer coerce environ to uppercase
- execve() on Windows now remotely controls its parent process to make
them spawn a replacement for itself. Then it'll be able to terminate
immediately once the spawn succeeds, without having to linger around
for the lifetime as a shell process for proxying the exit code. When
process worker thread running in the parent sees the child die, it's
given a handle to the new child, to replace it in the process table.
- execve() and posix_spawn() on Windows will now provide CreateProcess
an explicit handle list. This allows us to remove handle locks which
enables better fork/spawn concurrency, with seriously correct thread
safety. Other codebases like Go use the same technique. On the other
hand fork() still favors the conventional WIN32 inheritence approach
which can be a little bit messy, but is *controlled* by guaranteeing
perfectly clean slates at both the spawning and execution boundaries
- sigset_t is now 64 bits. Having it be 128 bits was a mistake because
there's no reason to use that and it's only supported by FreeBSD. By
using the system word size, signal mask manipulation on Windows goes
very fast. Furthermore @asyncsignalsafe funcs have been rewritten on
Windows to take advantage of signal masking, now that it's much more
pleasant to use.
- All the overlapped i/o code on Windows has been rewritten for pretty
good signal and cancelation safety. We're now able to ensure overlap
data structures are cleaned up so long as you don't longjmp() out of
out of a signal handler that interrupted an i/o operation. Latencies
are also improved thanks to the removal of lots of "busy wait" code.
Waits should be optimal for everything except poll(), which shall be
the last and final demon we slay in the win32 i/o horror show.
- getrusage() on Windows is now able to report RUSAGE_CHILDREN as well
as RUSAGE_SELF, thanks to aggregation in the process manager thread.
2023-10-08 12:36:18 +00:00
|
|
|
* @cancelationpoint
|
2022-03-25 14:11:44 +00:00
|
|
|
* @restartable
|
2020-06-15 14:18:57 +00:00
|
|
|
*/
|
|
|
|
ssize_t readv(int fd, const struct iovec *iov, int iovlen) {
|
2022-04-18 15:54:42 +00:00
|
|
|
ssize_t rc;
|
Make improvements
- We now serialize the file descriptor table when spawning / executing
processes on Windows. This means you can now inherit more stuff than
just standard i/o. It's needed by bash, which duplicates the console
to file descriptor #255. We also now do a better job serializing the
environment variables, so you're less likely to encounter E2BIG when
using your bash shell. We also no longer coerce environ to uppercase
- execve() on Windows now remotely controls its parent process to make
them spawn a replacement for itself. Then it'll be able to terminate
immediately once the spawn succeeds, without having to linger around
for the lifetime as a shell process for proxying the exit code. When
process worker thread running in the parent sees the child die, it's
given a handle to the new child, to replace it in the process table.
- execve() and posix_spawn() on Windows will now provide CreateProcess
an explicit handle list. This allows us to remove handle locks which
enables better fork/spawn concurrency, with seriously correct thread
safety. Other codebases like Go use the same technique. On the other
hand fork() still favors the conventional WIN32 inheritence approach
which can be a little bit messy, but is *controlled* by guaranteeing
perfectly clean slates at both the spawning and execution boundaries
- sigset_t is now 64 bits. Having it be 128 bits was a mistake because
there's no reason to use that and it's only supported by FreeBSD. By
using the system word size, signal mask manipulation on Windows goes
very fast. Furthermore @asyncsignalsafe funcs have been rewritten on
Windows to take advantage of signal masking, now that it's much more
pleasant to use.
- All the overlapped i/o code on Windows has been rewritten for pretty
good signal and cancelation safety. We're now able to ensure overlap
data structures are cleaned up so long as you don't longjmp() out of
out of a signal handler that interrupted an i/o operation. Latencies
are also improved thanks to the removal of lots of "busy wait" code.
Waits should be optimal for everything except poll(), which shall be
the last and final demon we slay in the win32 i/o horror show.
- getrusage() on Windows is now able to report RUSAGE_CHILDREN as well
as RUSAGE_SELF, thanks to aggregation in the process manager thread.
2023-10-08 12:36:18 +00:00
|
|
|
BEGIN_CANCELATION_POINT;
|
2024-05-26 23:53:13 +00:00
|
|
|
rc = readv_impl(fd, iov, iovlen);
|
Make improvements
- We now serialize the file descriptor table when spawning / executing
processes on Windows. This means you can now inherit more stuff than
just standard i/o. It's needed by bash, which duplicates the console
to file descriptor #255. We also now do a better job serializing the
environment variables, so you're less likely to encounter E2BIG when
using your bash shell. We also no longer coerce environ to uppercase
- execve() on Windows now remotely controls its parent process to make
them spawn a replacement for itself. Then it'll be able to terminate
immediately once the spawn succeeds, without having to linger around
for the lifetime as a shell process for proxying the exit code. When
process worker thread running in the parent sees the child die, it's
given a handle to the new child, to replace it in the process table.
- execve() and posix_spawn() on Windows will now provide CreateProcess
an explicit handle list. This allows us to remove handle locks which
enables better fork/spawn concurrency, with seriously correct thread
safety. Other codebases like Go use the same technique. On the other
hand fork() still favors the conventional WIN32 inheritence approach
which can be a little bit messy, but is *controlled* by guaranteeing
perfectly clean slates at both the spawning and execution boundaries
- sigset_t is now 64 bits. Having it be 128 bits was a mistake because
there's no reason to use that and it's only supported by FreeBSD. By
using the system word size, signal mask manipulation on Windows goes
very fast. Furthermore @asyncsignalsafe funcs have been rewritten on
Windows to take advantage of signal masking, now that it's much more
pleasant to use.
- All the overlapped i/o code on Windows has been rewritten for pretty
good signal and cancelation safety. We're now able to ensure overlap
data structures are cleaned up so long as you don't longjmp() out of
out of a signal handler that interrupted an i/o operation. Latencies
are also improved thanks to the removal of lots of "busy wait" code.
Waits should be optimal for everything except poll(), which shall be
the last and final demon we slay in the win32 i/o horror show.
- getrusage() on Windows is now able to report RUSAGE_CHILDREN as well
as RUSAGE_SELF, thanks to aggregation in the process manager thread.
2023-10-08 12:36:18 +00:00
|
|
|
END_CANCELATION_POINT;
|
2022-10-12 04:06:27 +00:00
|
|
|
STRACE("readv(%d, [%s], %d) → %'ld% m", fd, DescribeIovec(rc, iov, iovlen),
|
|
|
|
iovlen, rc);
|
2022-03-23 13:31:55 +00:00
|
|
|
return rc;
|
2020-06-15 14:18:57 +00:00
|
|
|
}
|