vfork() would not save the registers that the parent would expect
to have restored after execl() completed.
Specially that execl() would call execve() underneath, further messing
up the stack of the parent.
To avoid that, we fork() for now. Later we will revisit and cleanup
vfork()+execve() to actually have vfork() store all the register
that the parent expects to see, and not those left by execve().
In the original code, looking at the registers just before the call to
popen(), and after the call showed that r7 would get clobbered.
This would leave the caller with an invalid pointer, leading to all
kinds of data corruptions.
execve() is simpler that execl() in this case.
Bug: 5336252
Change-Id: I3bf718c0bb4c0439f6f2753f153cdea14175be9c
bionic has an i386 ifdef guard for the O_LARGEFILE flag. The reason for this
is not traceable in the git history, but it's most likely due to some old
toolchain or emulator problem.
Bug Id: 10035
Change-Id: I742ee6e50220276c2cb268d66f89b5655ef689fc
Signed-off-by: Dragos Tatulea <dragos.tatulea@intel.com>
Quote from Linux Programmer's Manual:
"If t is non-NULL, the return value is also stored in the memory
pointed to by t."
Change-Id: I8cb66b67e5f34c536ce2f0db76a6dc337c42ea3f
Signed-off-by: Jim Huang <jserv@0xlab.org>
* commit 'c59b248241b18492d20e112230a907c14400275b':
Use sigemptyset to initialise sigset_t values. sigset_t may not be implemented as an integral type.
sigset_t may not be implemented as an integral type.
Change-Id: I3f5ec86d6383685979d8d23ef508b6c609569b38
Signed-off-by: Raghu Gandham <raghu@mips.com>
This patch improves the handling of 64-bit parameters in syscalls on ARM.
The ARM EABI mandates that 64-bit quantities be passed in even/odd register
pairs, which requires special treatment.
This allows us to simplify our implementations of pread() and pwrite()
and remove the C stubs for pread64() and pwrite64().
Also add ftruncate64() to <unistd.h>
Change-Id: I407e2fd223ba0093dd2d0b04c6152fadfc9ce3ef
Bug 3107933
64-bit pread() and pwrite() is needed for ZipFileRO to be able to read
ludicrously large ZIP files just in case someone is crazy enough to do
it.
Also fix a license header that was apparently mangled.
Change-Id: I6819ef8b36e46b63b40749c95717b1ecf9307033
Merge commit 'ff7d94530db60cf4fe4a4c287ee8821ebdf6263a'
* commit 'ff7d94530db60cf4fe4a4c287ee8821ebdf6263a':
Hide the symbol of helper function __libc_android_abort
Merge commit 'f67e5211e045af1b12f646448a5a35f96ba5e8f1' into gingerbread-plus-aosp
* commit 'f67e5211e045af1b12f646448a5a35f96ba5e8f1':
Hide the symbol of helper function __libc_android_abort
Ideally __libc_android_abort would be static, but it could not be
because gcc would not allow calling a static function from an asm
statement. Instead, using GCC visibility is work around.
Change-Id: Ifff6b9957ca3f0fc03c75c3e42582a48d43cefa2
This also allows us to optimize the case where we increment an
uncontended semaphore (no need to call futex_wake() then).
Change-Id: Iad48efe8551dc66dc89d3e3f18c001e5a6c1939f
Also add missing declarations to misc. functions.
Fix clearerr() implementation (previous was broken).
Handle feature test macros like _POSIX_C_SOURCE properly.
Change-Id: Icdc973a6b9d550a166fc2545f727ea837fe800c4
When calling sysconf with _SC_NPROCESSORS_ONLN, the value one (1) was
returned on systems with two or more cores, since '/proc/stat' was
incorrectly parsed.
The function line_parser_getc (LineParser* p) read 128 characters of
input for each invocation.
The proper and probably aimed for behavior is to read 128 characters
at the first call, then for each subsequent call only return the next
buffered character until a new read is needed and only then read
another 128 characters.
Due to a flipped comparison between the two variables in_len and
in_pos that track the number of bytes of data read into the input
buffer and how much of it has been parsed, a new group of 128
characters were read at almost every call to line_parser_getc,
overwriting the still unhandled bytes from the previous call to
read. This caused the lines to be read to be sampled more than parsed.
Change-Id: I93eec3c8c9b9f19ef798748579d0977111b5c0bb
Signed-off-by: Christian Bejram <christian.bejram@stericsson.com>
The problem was that the 'defdname' field of res_state structure
was not properly initialized in __res_vinit(). This field is used
to store the default domain name, which is normally build from
calling gethostname() (see line 549 of res_init.c).
Unfortunately, in the typical Android case, gethostname() returns
an error (the hostname is configured) and a random stack string is
used later to build the DNS search list (see lines 556+ in res_init.c)
For the sake of illustration, let's say the search list is set to
a random value like 'xWLK'.
The end result is that when trying to result an unknown domain name
(e.g. 'www.ptn'), the query fails then the resolver tries to make a
new query with the DNS search list path(s) appended (e.g. 'www.ptn.xWLK').
The patch simply initializes 'defdname' to an empty string to avoid
this when the net.dns.search system property is not set.
Also contains whitespace/formatting fixes
Any of the setuid functions now updates /acct/uid/ with its own tid
before changing users. This is so we can properly account for cpu time
per uid.
Change-Id: I34186cf4d5228cac8439e582a9e26c01ef3011e4
Signed-off-by: Mike Chan <mike@android.com>
On ARM EABI, 64-bit function parameters must be aligned
to an even/odd register pair. The weird way these stubs
were written (using separate lo/hi parameters) prevented
this alignment from being enforced by the compiler.
The gHardy man pages specify the return type of ptsname_r to be char*, but the
return value to be 0 on success, negative on error and the gHardy stdlib.h
defines extern int ptsname_r(...).
Busybox telnetd fails to run successfully without this change.
Merge commit '5f32207a3db0bea3ca1c7f4b2b563c11b895f276' into eclair-mr2-plus-aosp
* commit '5f32207a3db0bea3ca1c7f4b2b563c11b895f276':
Wrap ARM abort() to improve stack trace.
The code generated for Thumb and Thumb2 targets has different handling
for abort(). Because abort() is "noreturn", it doesn't need to preserve
the callee-save registers. The Thumb2 version trashes LR and makes it
impossible to figure out who called abort().
This inserts a trivial stub function; net effect is stack traces are
reasonable after an abort().
For bug 2191452.
