]> sourceware.org Git - newlib-cygwin.git/log
newlib-cygwin.git
4 years agoOnly pass the minimum number of syscall arguments
Georg Sauthoff [Mon, 10 Feb 2020 20:19:58 +0000 (21:19 +0100)]
Only pass the minimum number of syscall arguments

Previously, __internal_syscall() compiled into asm-code that unconditionally
sets the syscall argument registers a0 to a5.

For example, the instruction sequence for a exit syscall looked like
this:

    li    a0, 1   # in ther caller of exit()
    # ...         # in newlib:
    li    a1, 0   # unused arguments
    li    a2, 0
    li    a3, 0
    li    a4, 0
    li    a5, 0
    li    a7, 93  # exit syscall number

(i.e. the binary contains then 5 superfluous instructions for this
one argument syscall)

This commit changes the RISC-V syscall code such that only the required
syscall argument registers are set.

GCC detects that argc is known at compile time and thus evaluates all the
if-statements where argc is used at compile time (tested with -O2 and -Os).

4 years agoCygwin: pty: Prevent potential errno overwriting.
Takashi Yano [Mon, 10 Feb 2020 11:42:45 +0000 (20:42 +0900)]
Cygwin: pty: Prevent potential errno overwriting.

- In push_to_pcon_screenbuffer(), open() and ioctl() are called.
  Since push_to_pcon_screenbuffer() is called in read() and write(),
  errno which is set in read() and write() code may be overwritten
  in open() or ioctl() call. This patch prevent this situation.

4 years agoCygwin: pty: Fix state mismatch caused in mintty.
Takashi Yano [Sun, 9 Feb 2020 14:47:30 +0000 (23:47 +0900)]
Cygwin: pty: Fix state mismatch caused in mintty.

- PTY has a bug reported in:
  https://cygwin.com/ml/cygwin/2020-02/msg00067.html.
  This is the result of state mismatch between real pseudo console
  attaching state and state variable. This patch fixes the issue.

4 years agoCygwin: pty: Inherit typeahead data between two input pipes.
Takashi Yano [Sun, 9 Feb 2020 14:46:59 +0000 (23:46 +0900)]
Cygwin: pty: Inherit typeahead data between two input pipes.

- PTY has a problem that the key input, which is typed during
  windows native app is running, disappear when it returns to shell.
  (Problem 3 in https://cygwin.com/ml/cygwin/2020-02/msg00007.html)
  This is beacuse pty has two input pipes, one is for cygwin apps
  and the other one is for native windows apps. The key input during
  windows native program is running is sent to the second input pipe
  while cygwin shell reads input from the first input pipe.
  This patch realize transfering input data between these two pipes.

4 years agoCygwin: pty: Add missing member initialization for struct pipe_reply.
Takashi Yano [Sun, 9 Feb 2020 14:46:03 +0000 (23:46 +0900)]
Cygwin: pty: Add missing member initialization for struct pipe_reply.

- For pseudo console support, struct pipe_reply was changed in the
  past, however, the initialization was not fixed.

4 years agoCygwin: pty: Remove debug codes and organize related codes.
Takashi Yano [Sun, 9 Feb 2020 14:46:02 +0000 (23:46 +0900)]
Cygwin: pty: Remove debug codes and organize related codes.

- Debug codes used in the early stage of pseudo console support are
  removed. (Regarding ALWAYS_USE_PCON and USE_API_HOOK) Along with
  this, the codes related to this change are organized.

4 years agoCygwin: pty: Avoid screen distortion on slave read.
Takashi Yano [Sun, 9 Feb 2020 14:46:01 +0000 (23:46 +0900)]
Cygwin: pty: Avoid screen distortion on slave read.

- Echo back print is distorted when the cygwin program which calls
  Win32 console API directly calls slave read(). This patch fixes
  the issue.

4 years agoCygwin: pty: Define mask_switch_to_pcon_in() in fhandler_tty.cc.
Takashi Yano [Sun, 9 Feb 2020 14:46:00 +0000 (23:46 +0900)]
Cygwin: pty: Define mask_switch_to_pcon_in() in fhandler_tty.cc.

- This patch moves the definition of mask_switch_to_pcon() from
  fhandler.h to fhandler_tty.cc.

4 years agoCygwin: pty: Use pinfo() rather than kill() with signal 0.
Takashi Yano [Thu, 6 Feb 2020 10:48:17 +0000 (19:48 +0900)]
Cygwin: pty: Use pinfo() rather than kill() with signal 0.

- PTY code has a problem that tcsh is terminated if the following
  command is executed.
    true; chcp &
  This seems to be caused by invalid pointer access which occurs
  when the process exits during the kill() code is execuetd. This
  patch avoids the issue by not using kill().

4 years agoTypo in license for newlib/libc/stdio/flags.c
Keith Packard [Wed, 5 Feb 2020 06:48:19 +0000 (22:48 -0800)]
Typo in license for newlib/libc/stdio/flags.c

Fix spelling:

MERCHANT I BILITY -> MERCHANT A BILITY

Signed-off-by: Keith Packard <keithp@keithp.com>
4 years agoTypo in license terms for newlib/libm/common/log2.c
Keith Packard [Wed, 5 Feb 2020 06:48:18 +0000 (22:48 -0800)]
Typo in license terms for newlib/libm/common/log2.c

The closing quotes were in the wrong place

Signed-off-by: Keith Packard <keithp@keithp.com>
4 years agolibgloss: Fix lseek semihosting bug on nios2 and m68k
Sandra Loosemore [Wed, 5 Feb 2020 04:34:13 +0000 (21:34 -0700)]
libgloss: Fix lseek semihosting bug on nios2 and m68k

When off_t is 32 bits, the value needs to be sign-extended to 64 bits
before shifting right to extract the high-order word.  Previously
negative offsets were incorrectly encoded.

Signed-off-by: Sandra Loosemore <sandra@codesourcery.com>
4 years agoCygwin: pty: Remove meaningless pointer increment.
Takashi Yano [Tue, 4 Feb 2020 12:25:52 +0000 (21:25 +0900)]
Cygwin: pty: Remove meaningless pointer increment.

- Since commit 73742508fcd8e994450582c1b7296c709da66764, a pointer
  increment in master write code which has no effect was remaining.

4 years agoCygwin: console: Revise color setting codes in legacy console mode.
Takashi Yano [Sat, 1 Feb 2020 04:28:39 +0000 (13:28 +0900)]
Cygwin: console: Revise color setting codes in legacy console mode.

- With this patch, foreground color and background color are allowed
  to be set simultaneously by 24 bit color escape sequence such as
  ESC[38;2;0;0;255;48;2;128;128;0m in legacy console mode.

4 years agoCygwin: fhandler_fifo.cc: add commentary
Ken Brown [Sat, 1 Feb 2020 21:36:31 +0000 (16:36 -0500)]
Cygwin: fhandler_fifo.cc: add commentary

4 years agoCygwin: fstat_helper: always use handle in call to get_file_attribute
Ken Brown [Thu, 30 Jan 2020 15:08:21 +0000 (10:08 -0500)]
Cygwin: fstat_helper: always use handle in call to get_file_attribute

When fhandler_base::fstat_helper is called, the handle h returned by
get_stat_handle() should be pc.handle() and should be safe to use for
getting the file information.  Previously, the call to
get_file_attribute() for FIFOs set the first argument to NULL instead
of h, thereby forcing the file to be opened for fetching the security
descriptor in get_file_sd().

4 years agoRISC-V: Use newlib nano specific libm.
Jim Wilson [Wed, 29 Jan 2020 22:46:17 +0000 (14:46 -0800)]
RISC-V: Use newlib nano specific libm.

The libm gamma functions use the _gamma_signgam field of the reentrant
structure, which changes offset with the --enable-newlib-reent-small
configure option, which means we need to use a newlib nano specific
version of libm in addition to libc in the nano.specs file.  Reported
by Keith Packard.  There is a riscv-gnu-toolchain patch that goes
along with this to create the new libm_nano.a file.

Signed-off-by: Jim Wilson <jimw@sifive.com>
4 years agoCygwin: document recent changes
Ken Brown [Wed, 29 Jan 2020 17:09:49 +0000 (12:09 -0500)]
Cygwin: document recent changes

4 years agoCygwin: AF_LOCAL: fix fcntl and dup if O_PATH is set
Ken Brown [Sat, 25 Jan 2020 18:08:00 +0000 (13:08 -0500)]
Cygwin: AF_LOCAL: fix fcntl and dup if O_PATH is set

Make fhandler_socket_local::dup and fhandler_socket_local::fcntl (a
new method) call fhandler_base::dup and fhandler_base::fcntl if O_PATH
is set.

We're viewing the socket as a disk file here, but there's no need to
implement the actions of fhandler_disk_file::dup and
fhandler_disk_file::fcntl, which do nothing useful in this case beyond
what the fhandler_base methods do.  (The extra actions are only useful
when I/O is going to be done on the file.)

4 years agoCygwin: AF_LOCAL::fstatvfs: use our handle if O_PATH is set
Ken Brown [Sat, 25 Jan 2020 12:45:10 +0000 (07:45 -0500)]
Cygwin: AF_LOCAL::fstatvfs: use our handle if O_PATH is set

If O_PATH is set, then the fhandler_socket_local object has a handle
that can be used for getting the statvfs information.  Use it by
calling fhandler_base::fstatvfs_by_handle.  Without this change,
fhandler_disk_file::fstatfvs would be called on a new fhandler_disk
object, which would then have to be opened.

4 years agoCygwin: AF_LOCAL: set appropriate errno on system calls
Ken Brown [Thu, 23 Jan 2020 20:11:15 +0000 (15:11 -0500)]
Cygwin: AF_LOCAL: set appropriate errno on system calls

If an AF_LOCAL socket is opened with O_PATH, all socket system calls
that take a file descriptor argument fail on the resulting descriptor.
Make sure that errno is set as on Linux for those calls that are
implemented on Linux.  In almost all cases it is ENOTSOCK.  There are
two exceptions:

- sockatatmark(3); errno is EBADF.

- bindresvport(3); errno is EAFNOSUPPORT if the second argument sin
  (of type struct sockaddr_in *) is non-NULL and satisfies
  sin->sin_family == AF_INET.

Finally, there are two BSD socket system calls implemented on Cygwin
but not Linux: getpeereid(3) and bindresvport_sa(3).  Set errno to
ENOTSOCK for these for consistency with the majority of the other
calls.

4 years agoCygwin: AF_LOCAL: allow opening with the O_PATH flag
Ken Brown [Thu, 23 Jan 2020 19:39:15 +0000 (14:39 -0500)]
Cygwin: AF_LOCAL: allow opening with the O_PATH flag

If that flag is not set, or if an attempt is made to open a different
type of socket, the errno is now EOPNOTSUPP instead of ENXIO.  This is
consistent with POSIX, starting with the 2016 edition.  Earlier
editions were silent on this issue.

Opening is done in a (new) fhandler_socket_local::open method by
calling fhandler_base::open_fs.

Also add a corresponding fhandler_socket_local::close method.

4 years agoUse remove-advertising-clause script to edit BSD licenses
Keith Packard [Tue, 28 Jan 2020 18:54:53 +0000 (10:54 -0800)]
Use remove-advertising-clause script to edit BSD licenses

This edits licenses held by Berkeley and NetBSD, both of which
have removed the advertising requirement from their licenses.

Signed-off-by: Keith Packard <keithp@keithp.com>
4 years agoRevert "newlib: fix fseek optimization with SEEK_CUR"
Corinna Vinschen [Wed, 29 Jan 2020 17:47:33 +0000 (18:47 +0100)]
Revert "newlib: fix fseek optimization with SEEK_CUR"

This reverts commit 59362c80e3a02c011fd0ef3d7f07a20098d2a9d5.

This breaks gnulib's autoconf test for POSIX compatibility of
fflush/fseek.  After fflush/fseek, ftello and lseek are out of
sync, with lseek having the wrong offset.  This breaks backward
compatibility with Cygwin applications.

Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
4 years agoCygwin: stat: fix st_mode of fifos
Corinna Vinschen [Wed, 29 Jan 2020 14:14:05 +0000 (15:14 +0100)]
Cygwin: stat: fix st_mode of fifos

Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
4 years agoCygwin: remove CYGWIN=dos_file_warning option
Corinna Vinschen [Tue, 28 Jan 2020 20:59:22 +0000 (21:59 +0100)]
Cygwin: remove CYGWIN=dos_file_warning option

This option has been disabled long ago and nobody missed it.
Removing drops a bit of unneeded code

Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
4 years agoCygwin: FIFO: fstatvfs: use our handle if O_PATH is set
Ken Brown [Fri, 24 Jan 2020 20:09:03 +0000 (15:09 -0500)]
Cygwin: FIFO: fstatvfs: use our handle if O_PATH is set

If O_PATH is set, then the fhandler_fifo object has a handle that can
be used for getting the statvfs information.  Use it by calling
fhandler_base::fstatvfs_by_handle.  Before this change,
fhandler_disk_file::fstatfvs was called on a new fhandler_disk_file
object, which would then have to be opened.

4 years agoCygwin: fhandler_disk_file::fstatvfs: refactor
Ken Brown [Fri, 24 Jan 2020 21:02:57 +0000 (16:02 -0500)]
Cygwin: fhandler_disk_file::fstatvfs: refactor

Define a new method fhandler_base::fstatvfs_by_handle, extracted from
fhandler_disk_file::fstatvfs, which gets the statvfs information when
a handle is available.

This will be used in future commits for special files that have been
opened with O_PATH.

4 years agoCygwin: fhandler_base::fstat_fs: accomodate the O_PATH flag
Ken Brown [Fri, 24 Jan 2020 19:54:47 +0000 (14:54 -0500)]
Cygwin: fhandler_base::fstat_fs: accomodate the O_PATH flag

Treat a special file opened with O_PATH the same as a regular file,
i.e., use its handle to get the stat information.

Before this change, fstat_fs opened the file a second time, with the
wrong flags and without closing the existing handle.  A side effect
was to change the openflags of the file, possibly causing further
system calls to fail.

Currently this change only affects FIFOs, but it will affect
AF_LOCAL/AF_UNIX sockets too once they support O_PATH.

4 years agoCygwin: symlink/mknod: fix ACL handling
Corinna Vinschen [Tue, 28 Jan 2020 16:57:50 +0000 (17:57 +0100)]
Cygwin: symlink/mknod: fix ACL handling

mknod32 actually creates a path_conv, just to call mknod_worker
with a win32 path.  This doesn't only require to create path_conv
twice, it also breaks permissions on filesystems supporting ACLs.

Fix this by passing the path_conv created in the caller down to
symlink_worker.  Also, while at it, simplify the handling of trailing
slashes and move it out of symlink_worker.  Especially use the
new PC_SYM_NOFOLLOW_DIR flag to avoid fiddeling with creating
a new path copy without the trailing slash.

Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
4 years agoCygwin: path_conv: add PC_SYM_NOFOLLOW_DIR flag
Corinna Vinschen [Tue, 28 Jan 2020 16:40:40 +0000 (17:40 +0100)]
Cygwin: path_conv: add PC_SYM_NOFOLLOW_DIR flag

Usually a trailing slash requires to follow an existing symlink,
even with PC_SYM_NOFOLLOW.  The reason is that "foo/" is equivalent
to "foo/." so the symlink is in fact not the last path component,
"." is.  This is default for almost all scenarios.

PC_SYM_NOFOLLOW_DIR now allows the caller to request not to
follow the symlink even if a trailing slash is given.  This can
be used in callers to perform certain functions Linux-compatible.

Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
4 years agoCygwin: add short comments to path_conv options
Corinna Vinschen [Tue, 28 Jan 2020 15:10:13 +0000 (16:10 +0100)]
Cygwin: add short comments to path_conv options

Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
4 years agoCygwin: drop __stdcall from close_all_files
Corinna Vinschen [Tue, 28 Jan 2020 14:33:53 +0000 (15:33 +0100)]
Cygwin: drop __stdcall from close_all_files

Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
4 years agoCygwin: move chmod_device declaration to winsup.h
Corinna Vinschen [Tue, 28 Jan 2020 14:33:05 +0000 (15:33 +0100)]
Cygwin: move chmod_device declaration to winsup.h

Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
4 years agoCygwin: pty: Revise code waiting for forwarding again.
Takashi Yano [Mon, 27 Jan 2020 11:22:24 +0000 (20:22 +0900)]
Cygwin: pty: Revise code waiting for forwarding again.

- After commit 6cc299f0e20e4b76f7dbab5ea8c296ffa4859b62, outputs of
  cygwin programs which call both printf() and WriteConsole() are
  frequently distorted. This patch fixes the issue.

4 years agoCygwin: console: Share readahead buffer within the same process.
Takashi Yano [Mon, 27 Jan 2020 12:14:32 +0000 (21:14 +0900)]
Cygwin: console: Share readahead buffer within the same process.

- The cause of the problem reported in
  https://www.cygwin.com/ml/cygwin/2020-01/msg00220.html is that the
  chars input before dup() cannot be read from the new file descriptor.
  This is because the readahead buffer (rabuf) in the console is newly
  created by dup(), and does not inherit from the parent. This patch
  fixes the issue.

4 years agoCygwin: FIFO: tweak fcntl and dup when O_PATH is set
Ken Brown [Thu, 23 Jan 2020 16:31:05 +0000 (16:31 +0000)]
Cygwin: FIFO: tweak fcntl and dup when O_PATH is set

fhandler_fifo::fcntl and fhandler_fifo::dup now call the corresponding
fhandler_base methods if the FIFO was opened with the O_PATH flag.

4 years agoCygwin: re-implement fhandler_fifo::open with O_PATH
Ken Brown [Thu, 23 Jan 2020 16:31:04 +0000 (16:31 +0000)]
Cygwin: re-implement fhandler_fifo::open with O_PATH

If the O_PATH flag is set, fhandler_fifo::open now simply calls
fhandler_base::open_fs.

The previous attempt to handle O_PATH in commit aa55d22c, "Cygwin:
honor the O_PATH flag when opening a FIFO", fixed a hang but otherwise
didn't do anything useful.

4 years agoCygwin: device_access_denied: return false if O_PATH is set
Ken Brown [Thu, 23 Jan 2020 16:31:04 +0000 (16:31 +0000)]
Cygwin: device_access_denied: return false if O_PATH is set

If O_PATH is set in the flags argument of
fhandler_base::device_access_denied, return false.  No
read/write/execute access should be required in this case.

Previously, the call to device_access_denied in open(2) would lead to
an attempt to open the file with read access even if the O_PATH flag
was set.

4 years agoCygwin: Bump DLL version to 3.1.3
Corinna Vinschen [Fri, 24 Jan 2020 09:17:31 +0000 (10:17 +0100)]
Cygwin: Bump DLL version to 3.1.3

Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
4 years agofhandler_proc.cc:format_proc_cpuinfo add rdpru flag
Brian Inglis [Thu, 23 Jan 2020 09:06:27 +0000 (02:06 -0700)]
fhandler_proc.cc:format_proc_cpuinfo add rdpru flag

rdpru flag is cpuid xfn 80000008 ebx bit 4 added in linux 5.5;
see AMD64 Architecture Programmer's Manual Volume 3:
General-Purpose and System Instructions
https://www.amd.com/system/files/TechDocs/24594.pdf#page=329
and elsewhere in that document

4 years agoCygwin: pty: Remove close() call just before reopening slave.
Takashi Yano [Thu, 23 Jan 2020 11:34:25 +0000 (20:34 +0900)]
Cygwin: pty: Remove close() call just before reopening slave.

- After commit da4ee7d60b9ff0bcdc081609a4467adb428d58e6, the issue
  reported in https://www.cygwin.com/ml/cygwin/2020-01/msg00209.html
  occurs. This patch fixes the issue.

4 years agoadd pseudo console fixes to release text newlib-3.3.0
Corinna Vinschen [Wed, 22 Jan 2020 10:05:51 +0000 (11:05 +0100)]
add pseudo console fixes to release text

Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
4 years agoCygwin: pty: Introduce disable_pcon in environment CYGWIN.
Takashi Yano [Tue, 21 Jan 2020 13:25:13 +0000 (22:25 +0900)]
Cygwin: pty: Introduce disable_pcon in environment CYGWIN.

- For programs which does not work properly with pseudo console,
  disable_pcon in environment CYGWIN is introduced. If disable_pcon
  is set, pseudo console support is disabled.

4 years agoCygwin: pty: Fix reopening slave in push_to_pcon_screenbuffer().
Takashi Yano [Tue, 21 Jan 2020 14:41:44 +0000 (23:41 +0900)]
Cygwin: pty: Fix reopening slave in push_to_pcon_screenbuffer().

- For programs compiled with -mwindows option, reopening slave is
  needed in push_to_pcon_screenbuffer(), however, it was not at
  appropriate place. This causes the problem reported in
  https://www.cygwin.com/ml/cygwin/2020-01/msg00161.html. This
  patch fixes the issue.

4 years agoBump up newlib release to 3.3.0
Jeff Johnston [Tue, 21 Jan 2020 20:17:43 +0000 (15:17 -0500)]
Bump up newlib release to 3.3.0

4 years agoChange the reent verify check option to document disabling it
Jeff Johnston [Tue, 21 Jan 2020 20:12:34 +0000 (15:12 -0500)]
Change the reent verify check option to document disabling it

- also change the handling of default_newlib_reent_check_verify to
  be the same as other default variables in configure.host
- regenerate newlib/configure

4 years agoDefault newlib_reent_check_verify to yes in configure.host
Jeff Johnston [Tue, 21 Jan 2020 19:54:49 +0000 (14:54 -0500)]
Default newlib_reent_check_verify to yes in configure.host

4 years agoCygwin: pty: Revise code waiting for forwarding by master_fwd_thread.
Takashi Yano [Tue, 21 Jan 2020 02:22:02 +0000 (11:22 +0900)]
Cygwin: pty: Revise code waiting for forwarding by master_fwd_thread.

- Though this rarely happens, sometimes the first printing of non-
  cygwin process does not displayed correctly. To fix this issue,
  the code for waiting for forwarding by master_fwd_thread is revised.

4 years agoriscv: Map between ieeefp.h exception bits and RISC-V FCSR bits
Keith Packard [Tue, 21 Jan 2020 06:46:36 +0000 (22:46 -0800)]
riscv: Map between ieeefp.h exception bits and RISC-V FCSR bits

If we had architecture-specific exception bits, we could just set them
to match the processor, but instead ieeefp.h is shared by all targets
so we need to map between the public values and the register contents.

Signed-off-by: Keith Packard <keithp@keithp.com>
4 years agoriscv: Add 'break' statements to fpsetround switch
Keith Packard [Tue, 21 Jan 2020 06:46:35 +0000 (22:46 -0800)]
riscv: Add 'break' statements to fpsetround switch

This makes the fpsetround function actually do something rather than
just return -1 due to the default 'fall-through' behavior of the switch
statement.

Signed-off-by: Keith Packard <keithp@keithp.com>
4 years agoriscv: Use current pseudo-instructions to access the FCSR register
Keith Packard [Tue, 21 Jan 2020 06:46:34 +0000 (22:46 -0800)]
riscv: Use current pseudo-instructions to access the FCSR register

Use fscsr and frcsr to store and read the FCSR register instead of
fssr and frsr.

Signed-off-by: Keith Packard <keithp@keithp.com>
4 years agoCygwin: document recent changes
Ken Brown [Fri, 17 Jan 2020 15:52:54 +0000 (10:52 -0500)]
Cygwin: document recent changes

4 years agoCygwin: fstatat, fchownat: support the AT_EMPTY_PATH flag
Ken Brown [Fri, 27 Dec 2019 22:38:32 +0000 (17:38 -0500)]
Cygwin: fstatat, fchownat: support the AT_EMPTY_PATH flag

Following Linux, allow the pathname argument to be an empty string if
the AT_EMPTY_PATH flag is specified.  In this case the dirfd argument
can refer to any type of file, not just a directory, and the call
operates on that file.  In particular, dirfd can refer to a symlink
that was opened with O_PATH | O_NOFOLLOW.

4 years agoCygwin: readlinkat: allow pathname to be empty
Ken Brown [Fri, 27 Dec 2019 22:17:35 +0000 (17:17 -0500)]
Cygwin: readlinkat: allow pathname to be empty

Following Linux, allow the pathname argument to be an empty string,
provided the dirfd argument refers to a symlink opened with
O_PATH | O_NOFOLLOW.  The readlinkat call then operates on that
symlink.

4 years agoCygwin: allow opening a symlink with O_PATH | O_NOFOLLOW
Ken Brown [Fri, 27 Dec 2019 16:43:58 +0000 (11:43 -0500)]
Cygwin: allow opening a symlink with O_PATH | O_NOFOLLOW

Up to now, opening a symlink with O_NOFOLLOW fails with ELOOP.
Following Linux, allow this to succeed if O_PATH is also specified.

4 years agoCygwin: normalize_win32_path: allow drive without trailing backslash
Ken Brown [Wed, 15 Jan 2020 15:49:21 +0000 (10:49 -0500)]
Cygwin: normalize_win32_path: allow drive without trailing backslash

Commit 283cb372, "Cygwin: normalize_win32_path: improve error
checking", required a prefix '\\?\' or '\??\' in the source path to be
followed by 'UNC\' or 'X:\', where X is a drive letter.  That was too
restrictive, since it disallowed the paths '\\?\X: and '\??\X:'.  This
caused problems when a user tried to use the root of a drive as the
Cygwin installation root, as reported here:

  https://cygwin.com/ml/cygwin/2020-01/msg00111.html

Modify the requirement so that '\??\X:' and '\\?\X:' are now allowed
as source paths, without a trailing backslash.

4 years agoCygwin: pty: Fix state mismatch caused in octave gui.
Takashi Yano [Thu, 16 Jan 2020 11:04:47 +0000 (20:04 +0900)]
Cygwin: pty: Fix state mismatch caused in octave gui.

- In octave gui, sometimes state mismatch between real pty state
  and state variable occurs. For example, this occurs when 'ls'
  command is executed in octave gui. This patch fixes the issue.

4 years agoCygwin: pty: Set console code page only if pseudo console is enabled.
Takashi Yano [Fri, 10 Jan 2020 11:47:12 +0000 (20:47 +0900)]
Cygwin: pty: Set console code page only if pseudo console is enabled.

- Input UTF-8 chars are garbled in ConEmu with cygwin connector if
  the environment does not support pseudo console. This patch fixes
  the issue.

4 years agoCygwin: pty: Disable FreeConsole() on close for non cygwin process.
Takashi Yano [Fri, 10 Jan 2020 11:46:26 +0000 (20:46 +0900)]
Cygwin: pty: Disable FreeConsole() on close for non cygwin process.

- After commit e1a0775dc0545b5f9c81b09a327fc110c538b7b4, the problem
  reported in https://www.cygwin.com/ml/cygwin/2020-01/msg00093.html
  occurs. For Gnu scren and tmux, calling FreeConsole() on pty close
  is necessary. However, if FreeConsole() is called, cygwin setup
  with '-h' option does not work. Therefore, the commit
  e1a0775dc0545b5f9c81b09a327fc110c538b7b4 delayed closing pty.
  This is the cause of the problem above. Now, instead of delaying
  pty close, FreeConsole() is not called if the process is non cygwin
  processes such as cygwin setup.

4 years agoCygwin: console: Add code to restore console mode on close.
Takashi Yano [Thu, 2 Jan 2020 13:17:16 +0000 (22:17 +0900)]
Cygwin: console: Add code to restore console mode on close.

- The console with 24bit color support has a problem that console
  mode is changed if cygwin process is executed in cmd.exe which
  started in cygwin shell. For example, cursor keys become not
  working if bash -> cmd -> true are executed in this order.
  This patch fixes the issue.

4 years agoCygwin: console: Add workaround for broken CSI3J in Win10 1809.
Takashi Yano [Tue, 14 Jan 2020 01:53:59 +0000 (10:53 +0900)]
Cygwin: console: Add workaround for broken CSI3J in Win10 1809.

- In Win10 1809, the cursor position sometimes goes out of screen
  by clear command in console. This seems to be caused by escape
  sequence CSI3J (ESC[3J). This happens only for 1809. This patch
  is a workaround for the issue.

4 years agoCygwin: console: Disable xterm mode for non cygwin process only.
Takashi Yano [Tue, 14 Jan 2020 02:09:42 +0000 (11:09 +0900)]
Cygwin: console: Disable xterm mode for non cygwin process only.

- Special function keys such as arrow keys or function keys do not
  work in ConEmu with cygwin-connector after commit
  6a06c6bc8f8492ea09aa3ae180fe94e4ac265611. This patch fixes the
  issue.

4 years agoCygwin: pty: Add missing CloseHandle() calls.
Takashi Yano [Tue, 14 Jan 2020 02:50:04 +0000 (11:50 +0900)]
Cygwin: pty: Add missing CloseHandle() calls.

- PTY code which support pseudo console has a problem that causes
  handle leaks. Four of these are bug in pty code, and the other
  one seems to be a bug of Windows10. ClosePseudoConsole() fails
  to close one internal handle. This patch fixes the issue.

4 years agoCygwin: pty: Fix the issue regarding open and close multiple PTYs.
Takashi Yano [Tue, 14 Jan 2020 04:10:54 +0000 (13:10 +0900)]
Cygwin: pty: Fix the issue regarding open and close multiple PTYs.

- If two PTYs are opened in the same process and the first one
  is closed, the helper process for the first PTY remains running.
  This patch fixes the issue.

4 years agoCygwin: select: Speed up select() call for pty, pipe and fifo.
Takashi Yano [Mon, 6 Jan 2020 14:38:34 +0000 (23:38 +0900)]
Cygwin: select: Speed up select() call for pty, pipe and fifo.

- The slowing down issue of X11 forwarding using ssh -Y, reported
  in https://www.cygwin.com/ml/cygwin/2019-12/msg00295.html,
  is due to the change of select() code for pty in the commit
  915fcd0ae8d83546ce135131cd25bf6795d97966. cygthread::detach()
  takes at most about 10msec because Sleep() is used in the thread.
  For this issue, this patch uses cygwait() instead of Sleep() and
  introduces an event to abort the wait. For not only pty, but pipe
  and fifo also have the same problem potentially, so this patch
  applies same strategy to them as well.

4 years agoCygwin: console: Make suspending process work properly.
Takashi Yano [Sun, 5 Jan 2020 13:25:55 +0000 (22:25 +0900)]
Cygwin: console: Make suspending process work properly.

- After commit f4b47827cf87f055687a0c52a3485d42b3e2b941, suspending
  process by Ctrl-Z does not work in console and results in hang up.
  This patch fixes the issue.

4 years agoCygwin: pty: Revise the code for setting code page of pseudo console.
Takashi Yano [Wed, 1 Jan 2020 06:50:36 +0000 (15:50 +0900)]
Cygwin: pty: Revise the code for setting code page of pseudo console.

- Fix the problem which overrides the code page setting, reported
  in https://www.cygwin.com/ml/cygwin/2019-12/msg00292.html.

4 years agoCygwin: pty: Remove destructor for fhandler_pty_master class.
Takashi Yano [Wed, 1 Jan 2020 06:49:41 +0000 (15:49 +0900)]
Cygwin: pty: Remove destructor for fhandler_pty_master class.

- The destructor for fhandler_pty_master class does not seem to be
  necessary anymore. Therefore, it has been removed.

4 years agoFixed crash on wine by adding NULL check after memchr
Arseniy Lartsev [Tue, 7 Jan 2020 15:34:39 +0000 (16:34 +0100)]
Fixed crash on wine by adding NULL check after memchr

This is not a joke, there are vendors out there who build software for cygwin
only. Besides, this NULL check is good to have anyway.

4 years agoCygwin: Add missing Linux #define of CPU_SETSIZE
Mark Geisert [Mon, 23 Dec 2019 06:45:54 +0000 (22:45 -0800)]
Cygwin: Add missing Linux #define of CPU_SETSIZE

Though our implementation of cpu sets doesn't need it, software from
Linux environments expects this definition to be present.  It's
documented on the Linux CPU_SET(3) man page but was left out due to
oversight.

Addresses https://cygwin.com/ml/cygwin/2019-12/msg00248.html

4 years agoPrevent more NULL ptr accesses due to Balloc out of memory
Jeff Johnston [Thu, 9 Jan 2020 20:18:14 +0000 (15:18 -0500)]
Prevent more NULL ptr accesses due to Balloc out of memory

- fix gdtoa-gethex.c, ldtoa.c, and strtodg.c to use eBalloc

4 years agoBump up release to 3.2.0 for yearly snapshot newlib-3.2.0 newlib-snapshot-20200102
Jeff Johnston [Thu, 2 Jan 2020 19:56:24 +0000 (14:56 -0500)]
Bump up release to 3.2.0 for yearly snapshot

4 years agodoc: add more details about adding documentation to HOWTO
Jon Turney [Wed, 18 Dec 2019 14:03:42 +0000 (14:03 +0000)]
doc: add more details about adding documentation to HOWTO

Add a little more detail to the checklist for adding documentation
Also update the list of supported sections

4 years agoCygwin: FIFO: use FILE_PIPE_REJECT_REMOTE_CLIENTS flag
Ken Brown [Sat, 21 Dec 2019 22:53:52 +0000 (17:53 -0500)]
Cygwin: FIFO: use FILE_PIPE_REJECT_REMOTE_CLIENTS flag

Add that flag to the pipe type argument when creating the Windows
named pipe.  And add a definition of that flag to ntdll.h (copied from
/usr/include/w32api/ddk/ntifs.h).

4 years agoCygwin: add 3.1.2 release text cygwin-3_1_2-release
Corinna Vinschen [Sat, 21 Dec 2019 10:44:08 +0000 (11:44 +0100)]
Cygwin: add 3.1.2 release text

Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
4 years agoOptimize setjmp/longjmp for moxie.
Anthony Green [Fri, 20 Dec 2019 14:00:26 +0000 (09:00 -0500)]
Optimize setjmp/longjmp for moxie.

We don't need to save/restore every register -- just those
we don't expect to be trashed by function calls.

4 years agoCygwin: pty: Fix ESC[?3h and ESC[?3l handling again.
Takashi Yano [Thu, 19 Dec 2019 11:03:30 +0000 (20:03 +0900)]
Cygwin: pty: Fix ESC[?3h and ESC[?3l handling again.

- Even with commit fe512b2b12a2cea8393d14f038dc3914b1bf3f60, pty
  still has a problem in ESC[?3h and ESC[?3l handling if invalid
  sequence such as ESC[?$ is sent. This patch fixes the issue.

4 years agoDon't display trailing '.' in _dcvt
Keith Packard [Wed, 18 Dec 2019 06:00:49 +0000 (22:00 -0800)]
Don't display trailing '.' in _dcvt

In the two helper functions that _dcvt calls for 'f' and 'e' mode, if
there are no digits to display after the decimal point, don't add one.

Signed-off-by: Keith Packard <keithp@keithp.com>
4 years agoFix gcvt to always show 'ndigits' of precision
Keith Packard [Wed, 18 Dec 2019 16:49:06 +0000 (08:49 -0800)]
Fix gcvt to always show 'ndigits' of precision

Leading zeros after the decimal point should not count
towards the 'ndigits' limit.

This makes gcvt match glibc and the posix gcvt man page.

Signed-off-by: Keith Packard <keithp@keithp.com>
4 years agoFix fcvt to only show 'ndigit' past decimal
Keith Packard [Wed, 18 Dec 2019 16:49:05 +0000 (08:49 -0800)]
Fix fcvt to only show 'ndigit' past decimal

Even if the number is really small and this means showing *no* digits.
This makes newlib match glibc, and the fcvt posix man page.

Signed-off-by: Keith Packard <keithp@keithp.com>
4 years agoCygwin: pty: Fix a bug regarding ESC[?3h and ESC[?3l handling.
Takashi Yano [Wed, 18 Dec 2019 16:07:33 +0000 (01:07 +0900)]
Cygwin: pty: Fix a bug regarding ESC[?3h and ESC[?3l handling.

- Midnight commander (mc) does not work after the commit
  1626569222066ee601f6c41b29efcc95202674b7 as reported in
  https://www.cygwin.com/ml/cygwin/2019-12/msg00173.html.
  This patch fixes the issue.

4 years agodoc: Untabify python scripts used for making man pages
Jon Turney [Mon, 16 Dec 2019 19:28:59 +0000 (19:28 +0000)]
doc: Untabify python scripts used for making man pages

These scripts fail with a TabError exception if 'python' is python3, as
mixing tabs and spaces in indentation is forbidden in python3.

4 years agoCygwin: Bump DLL version to 3.1.2
Corinna Vinschen [Wed, 18 Dec 2019 09:48:10 +0000 (10:48 +0100)]
Cygwin: Bump DLL version to 3.1.2

Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
4 years agoCygwin: add 3.1.1 release text belatedly
Corinna Vinschen [Wed, 18 Dec 2019 09:47:43 +0000 (10:47 +0100)]
Cygwin: add 3.1.1 release text belatedly

Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
4 years agoCygwin: autoload: Add missing GetProcessGroupAffinity/GetThreadGroupAffinity cygwin-3_1_1-release
Corinna Vinschen [Tue, 17 Dec 2019 15:44:54 +0000 (16:44 +0100)]
Cygwin: autoload: Add missing GetProcessGroupAffinity/GetThreadGroupAffinity

Both functions were introduce with Windows 7 only, so we need to
autoload them for the sake of Vista/2008.

Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
4 years agoSet __IEEE_LITTLE_ENDIAN for _XTENSA_EL__ (ESP32)
Keith Packard [Mon, 16 Dec 2019 21:56:11 +0000 (13:56 -0800)]
Set __IEEE_LITTLE_ENDIAN for _XTENSA_EL__ (ESP32)

Signed-off-by: Keith Packard <keithp@keithp.com>
4 years agoReturn EINVAL for illegal base in strtol
Keith Packard [Mon, 16 Dec 2019 21:55:30 +0000 (13:55 -0800)]
Return EINVAL for illegal base in strtol

Signed-off-by: Keith Packard <keithp@keithp.com>
4 years agoCygwin: Bump DLL version to 3.1.1
Corinna Vinschen [Mon, 16 Dec 2019 20:09:50 +0000 (21:09 +0100)]
Cygwin: Bump DLL version to 3.1.1

Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
4 years agoCygwin: add strtold fix to release message cygwin-3_1_0-release
Corinna Vinschen [Mon, 16 Dec 2019 15:39:17 +0000 (16:39 +0100)]
Cygwin: add strtold fix to release message

Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
4 years agostrtold: set errno to ERANGE on underflow per POSIX
Bruno Haible [Mon, 16 Dec 2019 14:11:47 +0000 (15:11 +0100)]
strtold: set errno to ERANGE on underflow per POSIX

https://pubs.opengroup.org/onlinepubs/9699919799/functions/strtod.html

4 years agoCygwin: ilogbl: Make sure to return FP_ILGB0 on zero input
Corinna Vinschen [Mon, 16 Dec 2019 09:50:17 +0000 (10:50 +0100)]
Cygwin: ilogbl: Make sure to return FP_ILGB0 on zero input

Signed-off-by: Corinna Vinschen <corinna@vinschen.de>
4 years agoMSP430: Support new msp430-elfbare target
Jozef Lawrynowicz [Mon, 25 Nov 2019 21:17:41 +0000 (21:17 +0000)]
MSP430: Support new msp430-elfbare target

Update the target triplet glob used when configuring for msp430 to
support a new msp430-elfbare target being added to gcc.

4 years agoImplement the unlink system call for the moxie simulator.
Anthony Green [Sat, 14 Dec 2019 10:27:38 +0000 (05:27 -0500)]
Implement the unlink system call for the moxie simulator.

Corresponding support for this was just added to the gdb moxie simulator.
Unlink support is required by the GCC testsuite.

4 years agoFix setjmp/longjmp for the moxie port.
Anthony Green [Fri, 13 Dec 2019 18:08:06 +0000 (13:08 -0500)]
Fix setjmp/longjmp for the moxie port.

These functions needs to save and restore the stack frame, because
that's where the return address is stored.

4 years agoCygwin: document the last bugfix
Ken Brown [Tue, 10 Dec 2019 13:45:17 +0000 (08:45 -0500)]
Cygwin: document the last bugfix

4 years agoCygwin: symlink_info::check: avoid assertion failure
Ken Brown [Mon, 9 Dec 2019 12:27:18 +0000 (07:27 -0500)]
Cygwin: symlink_info::check: avoid assertion failure

On certain error conditions there is a code snippet that checks
whether the last component of the path has a trailing dot or space or
a leading space.  Skip this check if the last component is empty,
i.e., if the path ends with a backslash.  This avoids an assertion
failure if the trailing backslash is the only backslash in the path,
as is the case for a DOS drive 'X:\'.

Addresses: https://cygwin.com/ml/cygwin/2019-12/msg00016.html

4 years agolibm: switch sf_log1p from double error routines to float
Keith Packard [Fri, 29 Nov 2019 19:23:20 +0000 (11:23 -0800)]
libm: switch sf_log1p from double error routines to float

sf_log1p was using __math_divzero and __math_invalid, which
drag in a pile of double-precision code. Switch to using the
single-precision variants. This also required making those
available in __OBSOLETE_MATH mode.

Signed-off-by: Keith Packard <keithp@keithp.com>
4 years agoCygwin: /proc/[PID]/stat to pull process priority correctly
Anton Lavrentiev via cygwin-patches [Sun, 1 Dec 2019 03:58:14 +0000 (22:58 -0500)]
Cygwin: /proc/[PID]/stat to pull process priority correctly

Fix to prior commit 5fa9a0e7 to address https://cygwin.com/ml/cygwin/2019-08/msg00082.html

4 years agonewlib/libc/include/sys/features.h: update __STDC_ISO_10646__
Brian Inglis [Tue, 26 Nov 2019 15:34:42 +0000 (08:34 -0700)]
newlib/libc/include/sys/features.h: update __STDC_ISO_10646__

newlib wide char conversion functions were updated to
Unicode 11 on 2019-01-12
update standard symbol __STDC_ISO_10646__ to
Unicode 11 release date 2018-06-05 for Cygwin

This page took 0.071211 seconds and 5 git commands to generate.