| Commit message (Collapse) | Author | Files | Lines |
|---|
|
* libc/gxx_wrappers.c: Modifications to make gcc 3.x happy.
|
|
* include/rtems/libio_.h: Added ifdef to ensure that LOGIN_NAME_MAX
is defined on GNU/Linux (RedHat 6.2 distribution).
* libc/unixlibc.c: Added stubs for rtems_io_register_name() and
rtems_io_lookup_name() so UNIX port would compile the IO manager
sptest that registers /dev/null.
|
|
* libc/lseek.c: Modified after discussion with Eugeny S. Mints
<jack@oktet.ru> to correct the behavior. There were two mistakes.
First, iop->offset was incorrectly set for SEEK_END. Second,
iop->offset should be left unmodified if there are errors.
This modification attempts to fix both situations.
|
|
* libc/lseek.c: Modified after discussion with Eugeny S. Mints
<jack@oktet.ru> to correct the behavior. There were two mistakes.
First, iop->offset was incorrectly set for SEEK_END. Second,
iop->offset should be left unmodified if there are errors.
This modification attempts to fix both situations.
|
|
* libc/termios.c: Fix a bug in the termios implementation in
the following scenario:
The General Terminal Interface document that me states that
if VMIN = 0 and VTIME = 0, then read() should return the minimum
of two values:
a) number of bytes available
b) number of bytes requested (I assume from the read call)
The current implementation of the fillBufferQueue() in termios.c is
always return 1 character with these setting values. I know the
termios buffer has more than one character available and my read()
call is requesting 1024 bytes.
|
|
* include/rtems/termiostypes.h, include/sys/ioccom.h:
Update of PPPD to 2.3.11 from 2.3.5 touched these files.
|
|
* libc/termios.c: Fix a bug in the termios implementation in
the following scenario:
The General Terminal Interface document that me states that
if VMIN = 0 and VTIME = 0, then read() should return the minimum
of two values:
a) number of bytes available
b) number of bytes requested (I assume from the read call)
The current implementation of the fillBufferQueue() in termios.c is
always return 1 character with these setting values. I know the
termios buffer has more than one character available and my read()
call is requesting 1024 bytes.
|
|
* libc/getgrent.c, libc/getpwent.c: the 'ls' and more related command
previous failed after a chroot(). (unknown user & group)
|
|
* libc/libio_sockets.c (rtems_bsdnet_fdToSocket): Per bug
report from Gene Smith <Gene.Smith@sea.siemens.com>, enhanced the
error checking to account for a socket being closed.
|
|
* libc/base_fs.c: Deleted chmod() now that IMFS creates the
root node with the desired permissions. chmod() is also not
supported by the miniIMFS so this is not allowable.
|
|
* include/rtems/libio_.h: Added login name, uid, gid, euid, and egid
so they can be part of the private environment.
|
|
* libc/ttyname.c (ttyname_r): Removed duplicate call to closedir().
* libc/getpwent.c: Create a more robust /etc/passwd and /etc/group.
* libc/base_fs.c: Change permissions of files and directories. Now uses
octal constants.
|
|
* libc/write.c: Change type of rc to match return type from
underlying function per report from Eric Norum.
|
|
* libc/libio.c (rtems_libio_allocate): Make sure size and offset
fields are cleared on each file open. Before this field was cleared,
this resulted in the value from the last time that IOP was used
being still in place. Discovered by Andrew Bythell
<abythell@nortelnetworks.com>.
* libc/open.c: Remove redundant setting of iop->offset.
|
|
* include/rtems/libio_.h, libc/chroot.c, libc/privateenv.c:
Private environment and chroot() enhancements and fixes. Comments:
+ privateenv has been modified to let at chroot() to be more
POSIX like Sergei Organov recommended.
+ A task owner lets that rtems_set_private_env() will be
called twice or more times.
+ chroot() can be called without a previous
rtems_set_private_env(); (transpanrently)
+ The second call of rtems_set_private_env() makes a internal
chroot("/") into global imfs_root.
+ chroot() runs like chdir() without a previous chdir("/") with
the global root.
+ The current directory can be in a wrong place like Linux and
many other Unices.
|
|
* include/rtc.h: New file.
* include/Makefile.am: Updated to reflect above.
|
|
* sys/utsname.h: Added C++ wrapper.
|
|
* libc/chroot.c: Fixed naming problems. Names were changed
when the code this depends on was merged.
|
|
* cpu.c, rtems/score/cpu.h: Bug report from Peter Mueller
<peter.o.mueller@gmx.de> because of not correcting for the ISR
vector table now being allocated from the workspace.
|
|
* Alternate email is correo@fernando-ruiz.com
* libc/privateenv.c: New file.
* include/rtems/libio_.h, libc/Makefile.am, libc/base_fs.c,
libc/unmount.c: Moved default umask, current working directory,
root, and links followed count into a structure "user environment"
that can then be treated as a unit. This enable giving unique
copies of these to individual threads or collection of threads.
|
|
* configure.in: Add src/config.h
* src/Makefile.am: Add INCLUDES += -I. to pickup config.h
* src/.cvsignore: Add config.h and stamp-h
* src/*.c: Add config.h support.
|
|
* configure.in: Add libc/config.h
* libc/Makefile.am: Add INCLUDES += -I. to pickup config.h
* libc/.cvsignore: Add config.h and stamp-h
* libc/*.c: Add config.h support.
|
|
* libc/mallocfreespace.c: Name of routine is
rtems_region_get_information() NOT region_get_information().
* libc/unmount.c: Removed unused variable.
|
|
* include/rtems/Makefile.am: Added termiostypes.h.
* libc/Makefile.am: Removed termiostypes.h.
* libc/termios.c: Changed include of "termiostypes.h" to
<rtems/termiostypes.h> since that is an RTEMS specific header file.
|
|
* include/sys/ttycom.h: New file -- moved from libnetworking/sys.
* Makefile.am: Modified to reflect above.
|
|
* libc/getcwd.c: Ensure _closedir(dir) is called even on
error exit from the routine.
|
|
* libc/termiostypes.h: New file -- missed in earlier commits.
|
|
* libc/termios.c, libc/termiostypes.h: Task driver driver model
and line discipline support from Thomas Doerfler
<Thomas.Doerfler@imd-systems.de>.
|
|
* libc/mount.c: Make sure there is space allocated for a device name
in the mount table entry.
|
|
* libc/mount.c (search_mt_for_mount_point): Deleted routine.
* libc/mount.c (Is_node_fs_root): Replacement for above that
accounts for the imaginary root node being returned by the
filesystem evaluation routine.
* libc/unmount.c (unmount): Account for imaginary root node
being returned and improved variable names to clarify code.
* libc/unmount.c (file_systems_below_this_mountpoint): Body of
routine replaced to account for imaginary root node being returned.
|
|
* libc/libio.c: Special treatment of O_NDELAY for Cygwin.
|
|
* Makefile.am: Use ... instead of RTEMS_TOPdir in ACLOCAL_AMFLAGS.
|
|
* Makefile.am: Switch to ACLOCAL_AMFLAGS = -I $(RTEMS_TOPdir)/aclocal.
|
|
* include/Makefile.am, include/rtems/libio_.h, libc/Makefile.am,
libc/assoc.c, libc/assocnamebad.c, libc/base_fs.c, libc/cfsetispeed.c,
libc/cfsetospeed.c, libc/chdir.c, libc/chmod.c, libc/chown.c,
libc/close.c, libc/closedir.c, libc/dup2.c, libc/error.c, libc/eval.c,
libc/fchdir.c, libc/fchmod.c, libc/fcntl.c, libc/fdatasync.c,
libc/fpathconf.c, libc/fstat.c, libc/fsync.c, libc/ftruncate.c,
libc/getdents.c, libc/ioctl.c, libc/libio.c, libc/libio_sockets.c,
libc/link.c, libc/lseek.c, libc/malloc.c, libc/mallocfreespace.c,
libc/mknod.c, libc/mount.c, libc/newlibc.c, libc/no_libc.c,
libc/open.c, libc/read.c, libc/readlink.c, libc/rmdir.c, libc/stat.c,
libc/symlink.c, libc/tcsetattr.c, libc/telldir.c, libc/ttyname.c,
libc/ttyname_r.c, libc/umask.c, libc/unlink.c, libc/unmount.c,
libc/utime.c, libc/write.c:
assoc.h, error.h, libio_.h, libio.h, and libcsupport.h moved
from libc to lib/include/rtems and now must be referenced as
<rtems/XXX.h>.
* include/rtems/Makefile.am, include/rtems/.cvsignore: New file.
* include/rtems/assoc.h, include/rtems/error.h,
include/rtems/libcsupport.h, include/rtems/libio.h,
include/rtems/libio_.h: New/moved files.
|
|
* configure.in: Remove bogus cfg_subdirs.
* Makefile.am: ACLOCAL_AMFLAGS= -I $(RTEMS_TOPdir)/macros.
Switch to GNU canonicalization.
|
|
* libc/ioman.c: Moved to libfs.
* libc/Makefile.am: Removed ioman.c as part of moving it to libfs.
* libc/base_fs.c: Removed include of imfs.h and reworded comment
to avoid being IMFS specific.
* libc/libio.h: Removed prototype of IMFS_ops since it should
not be in this file.
* libc/mount.c: Removed IMFS specific configuration information.
* libc/unmount.c: Removed include of imfs.h and reworded comment
to avoid being IMFS specific.
|
|
* libc/Makefile.am: Added mallocfreespace.c.
* libc/mallocfreespace.c: New file based on work by Nick Simon
<Nick.SIMON@syntegra.bt.co.uk> which he included in malloc.c.
* libc/libcsupport.h: Added prototype for malloc_free_space().
|
|
* libc/fcntl.c: Do not require every filesystem to have an fcntl()
handler. Most fcntl() operations can be performed with no
filesystem support.
* ChangeLog: Corrected comment.
|
|
* libc/libio.h (rtems_filesystem_file_handlers_r,
rtems_filesystem_operations_table): Added _h to all structure
fields to indicate they are "handlers".
* libc/libio_.h, libc/chdir.c, libc/chmod.c, libc/chown.c,
libc/close.c, libc/eval.c, libc/fchdir.c, libc/fchmod.c,
libc/fcntl.c, libc/fdatasync.c, libc/fstat.c, libc/fsync.c,
libc/ftruncate.c, libc/getdents.c, libc/imfs_eval.c,
libc/imfs_unlink.c, libc/ioctl.c, libc/ioman.c, libc/link.c,
libc/lseek.c, libc/mknod.c, libc/mount.c, libc/open.c, libc/read.c,
libc/readlink.c, libc/rmdir.c, libc/stat.c, libc/symlink.c,
libc/unlink.c, libc/unmount.c, libc/utime.c, libc/write.c:
Modified to reflect above name change.
|
|
* libc/newlibc.c: Removed _A29K specific ifdef.
|
|
* libio.c: Added O_NONBLOCK to list of flags.
|
|
* Many files: Moved posix/include/rtems/posix/seterr.h to
score/include/rtems/seterr.h so it would be available within
all APIs.
|
|
* configure.in: Add lib/libcpu and lib/libbsp to cfg_subdirs
|
|
* libc/fs_null_handlers.c: Added missing rmnod entry.
|
|
* libc/isatty.c, libc/imfs_handlers_directory.c, libc/creat.c,
libc/imfs_directory.c: Fixed style issues.
|
|
* libc/chmod.c: Return ENOTSUP if filesystem does not have handler.
* libc/eval.c: Ditto.
* libc/fcntl.c: Ditto.
* libc/fsync.c: Ditto.
* libc/ioctl.c: Ditto.
* libc/ioman.c: Ditto.
* libc/link.c: Ditto.
* libc/memfile.c: Ditto.
* libc/mknod.c: Ditto.
* libc/symlink.c: Ditto.
* libc/libio.h(rtems_filesystem_dev_major_t): New macro.
* libc/libio.h(rtems_filesystem_dev_minor_t): New macro.
|
|
====
-4 = {d = -4, i = {0xc0100000, 0x0}} s=1 e=0x80 m=0x100000 (0x0900000)
-3 = {d = -3, i = {0xc0080000, 0x0}} s=1 e=0x80 m=0x080000 (0x0880000)
-2 = {d = -2, i = {0xc0000000, 0x0}} s=1 e=0x80 m=0x000000 (0x0800000)
-1 = {d = -1, i = {0xbff00000, 0x0}} s=1 e=0x7f m=0x000000 (0x0800000)
0 = {d = 0, i = {0x00000000, 0x0}} s=0 e=0 m=0x000000 (0x0800000)
1 = {d = 1, i = {0x3ff00000, 0x0}} s=0 e=0x7f m=0x000000 (0x0800000)
2 = {d = 2, i = {0x40000000, 0x0}} s=0 e=0 m=0x000000 (0x0800000)
3 = {d = 3, i = {0x40080000, 0x0}} s=0 e=0 m=0x080000 (0x0880000)
C4X
===
-4 = {d = -4, i = {0x01800000, 0x0}} s=1 e=1 m=0x800000 (0x1000000)
-3 = {d = -3, i = {0x01c00000, 0x0}} s=1 e=1 m=0xa00000 (0x1400000)
-2 = {d = -2, i = {0x00800000, 0x0}} s=1 e=0 m=0x800000 (0x1000000)
-1 = {d = -1, i = {0xff800000, 0x0}} s=1 e=0xff m=0x000000 (0x1000000)
0 = {d = 0, i = {0x80000000, 0x0}} s=0 e=0 m=0x000000 (0x0800000)
1 = {d = 1, i = {0x00000000, 0x0}} s=0 e=0x7f m=0x000000 (0x0800000)
2 = {d = 2, i = {0x01000000, 0x0}} s=0 e=1 m=0x400000 (0x0800000)
3 = {d = 3, i = {0x01400000, 0x0}} s=0 e=1 m=0x600000 (0x0c00000)
4 = {d = 4, i = {0x02000000, 0x0}} s=0 e=2 m=0xXXXXXX (0x0000000)
0x1010 0000
0101 0000
0010 1002
|
|
the correct circumstances of DMA buffer size, serial
line interrupts, and ethernet interrupts the termios osend routine would
lock up waiting for the raw output buffer semaphore.
|
|
a BSP (c4xsim) supporting the simulator included with gdb. This port
was done by Joel Sherrill and Jennifer Averett of OAR Corporation.
Also included with this port is a space/time optimization to eliminate
FP context switch management on CPUs without hardware or software FP.
An issue with this port was that sizeof(unsigned32) = sizeof(unsigned8)
on this CPU. This required addressing alignment checks and assumptions
as well as fixing code that assumed sizeof(unsigned32) == 4.
|
|
functionality to libc and update TODO.
|
