| Age | Commit message (Collapse) | Author |
|---|
|
On some combinations of Montavista Linux running on Cavium Octeon
chips, some socket-related system calls erroneously return negative
numbers other than -1 to indicate errors, but inet_drv.c specifically
compares against -1 to test for errors. The result is that beam dumps
core due to the code treating these negative numbers as success
indicators, as counts/offsets of bytes written, etc. thereby
corrupting its own internal data structures.
To fix this, introduce a portability macro to test the result of
socket system calls. The test remains unchanged on Windows but for
other platforms the macro considers all return values that are less
than zero to be errors.
Though POSIX specifies that errors from these system calls are
indicated by a return value of -1, treating all negative return values
as errors is also safe, as described in detail below. In networking
programming, treating all negative return values from system calls as
errors is very common practice -- see the examples in W. Richard
Stevens's popular and highly lauded network programming books, for
example.
For system calls that return 0 to indicate success, treating all
negative numbers as errors is safe because only 0 is specified to
indicate success. These include:
getsockname
getpeername
getsockopt
gethostname
bind
listen
connect
close
shutdown
Likewise, for system calls that return non-negative numbers to
indicate success, treating all negative numbers as errors is also
safe. These functions typically return signed integers of type
ssize_t, and they treat any parameters of type size_t that cannot fit
within the ssize_t return value, such as numbers of bytes to read or
write, as errors (specifically EINVAL). For example, in the "ERRORS"
section of the man page for writev from several varieties of Linux, it
states that EINVAL is returned when the total length of the I/O is
more than can be expressed by the ssize_t return value.
These calls include:
recv
recvfrom
recvmsg
writev
send
sendto
sendmsg
Finaly, the socket() system call is also similar to these in that it
returns a signed type (int) with all non-negative return values
indicating success, so treating all negative return values as errors
is safe.
|
|
* fm/file-operations:
Update preloaded modules
Add file:advise/4 - a wrapper to the POSIX syscall posix_fadvise
Add file:datasync/1 for syncing file contents only
sys.h: Correct the get_int64() macro
OTP-8637 fm/file-operations
The functions file:advise/4 and file:datasync/1 have been added. (Thanks to
Filipe David Manana.)
|
|
Useful for informing the Operating System about the access pattern
for a file's data, so that it can adapt the caching strategy to
maximize disk IO performance.
|
|
file:datasync/1 invokes the POSIX system call "int fdatasync(int fd)".
This system call is similar to "fsync" but, unlike fsync, it does not
update the metadata associated with the file (like the access time for
example). It's used by many DBMSs (MySQL and SQLite of example) to
increase disk IO performance, as it avoids disk seeks and disk write
operations compared to fsync.
More details on it at:
http://linux.die.net/man/2/fdatasync
An example, from the MySQL source:
http://bazaar.launchpad.net/~mysql/mysql-server/mysql-5.1-telco-6.1/annotate/head%3A/mysys/my_sync.c#L61
This new function just calls fsync on systems not implementing fdatasync.
|
|
|
|
* bg/remove-stray-ose-support:
configure: Remove stray OSE/Delta support
Makefiles: Remove stray OSE/Delta support
kernel tests: Remove stray OSE/Delta support
system tests: Remove stray OSE/Delta support
erl_interface tests: Remove stray OSE/Delta support
epmd: Remove stray OSE/Delta support
epmd: #ifdef out start_epmd() for other platforms than VxWorks
emulator tests: Remove stray OSE/Delta support
emulator: Remove stray OSE/Delta support
emulator: Eliminate #ifdef for sys_tty_reset()
test_server: Remove stray support for OSE/Delta
OTP-8585 bg/remove-stray-ose-support
|
|
|
|
|
|
* pan/otp_8332_halfword:
Teach testcase in driver_suite the new prototype for driver_async
wx: Correct usage of driver callbacks from wx thread
Adopt the new (R13B04) Nif functionality to the halfword codebase
Support monitoring and demonitoring from driver threads
Fix further test-suite problems
Correct the VM to work for more test suites
Teach {wordsize,internal|external} to system_info/1
Make tracing and distribution work
Turn on instruction packing in the loader and virtual machine
Add the BeamInstr data type for loaded BEAM code
Fix the BEAM dissambler for the half-word emulator
Store pointers to heap data in 32-bit words
Add a custom mmap wrapper to force heaps into the lower address range
Fit all heap data into the 32-bit address range
|
|
Store Erlang terms in 32-bit entities on the heap, expanding the
pointers to 64-bit when needed. This works because all terms are stored
on addresses in the 32-bit address range (the 32 most significant bits
of pointers to term data are always 0).
Introduce a new datatype called UWord (along with its companion SWord),
which is an integer having the exact same size as the machine word
(a void *), but might be larger than Eterm/Uint.
Store code as machine words, as the instructions are pointers to
executable code which might reside outside the 32-bit address range.
Continuation pointers are stored on the 32-bit stack and hence must
point to addresses in the low range, which means that loaded beam code
much be placed in the low 32-bit address range (but, as said earlier,
the instructions themselves are full words).
No Erlang term data can be stored on C stacks (enforced by an
earlier commit).
This version gives a prompt, but test cases still fail (and dump core).
The loader (and emulator loop) has instruction packing disabled.
The main issues has been in rewriting loader and actual virtual
machine. Subsystems (like distribution) does not work yet.
|
|
|
|
It must be unsigned so that prim_inet will not reject
when it is sent down again.
(Suggested fix by Raimo for a bug reported by Simon Cornish.)
|
|
|
