- 27 9月, 2011 2 次提交
-
-
由 Rich Felker 提交于
lock out new waiters during the broadcast. otherwise the wait count added to the mutex might be lower than the actual number of waiters moved, and wakeups may be lost. this issue could also be solved by temporarily setting the mutex waiter count higher than any possible real count, then relying on the kernel to tell us how many waiters were requeued, and updating the counts afterwards. however the logic is more complex, and i don't really trust the kernel. the solution here is also nice in that it replaces some atomic cas loops with simple non-atomic ops under lock.
-
由 Rich Felker 提交于
due to moving waiters from the cond var to the mutex in bcast, these waiters upon wakeup would steal slots in the count from newer waiters that had not yet been signaled, preventing the signal function from taking any action. to solve the problem, we simply use two separate waiter counts, and so that the original "total" waiters count is undisturbed by broadcast and still available for signal.
-
- 26 9月, 2011 4 次提交
-
-
由 Rich Felker 提交于
the changes to syscall_ret are mostly no-ops in the generated code, just cleanup of type issues and removal of some implementation-defined behavior. the one exception is the change in the comparison value, which is fixed so that 0xf...f000 (which in principle could be a valid return value for mmap, although probably never in reality) is not treated as an error return.
-
由 Rich Felker 提交于
testing revealed that the old implementation, while correct, was giving way too many spurious wakeups due to races changing the value of the condition futex. in a test program with 5 threads receiving broadcast signals, the number of returns from pthread_cond_wait was roughly 3 times what it should have been (2 spurious wakeups for every legitimate wakeup). moreover, the magnitude of this effect seems to grow with the number of threads. the old implementation may also have had some nasty race conditions with reuse of the cond var with a new mutex. the new implementation is based on incrementing a sequence number with each signal event. this sequence number has nothing to do with the number of threads intended to be woken; it's only used to provide a value for the futex wait to avoid deadlock. in theory there is a danger of race conditions due to the value wrapping around after 2^32 signals. it would be nice to eliminate that, if there's a way. testing showed no spurious wakeups (though they are of course possible) with the new implementation, as well as slightly improved performance.
-
由 Rich Felker 提交于
using swap has a race condition: the waiters must be added to the mutex waiter count *before* they are taken off the cond var waiter count, or wake events can be lost.
-
由 Rich Felker 提交于
-
- 25 9月, 2011 2 次提交
-
-
由 Rich Felker 提交于
somehow i forgot that normal-type mutexes don't store the owner tid.
-
由 Rich Felker 提交于
this avoids the "stampede effect" where pthread_cond_broadcast would result in all waiters waking up simultaneously, only to immediately contend for the mutex and go back to sleep.
-
- 24 9月, 2011 3 次提交
-
-
由 Rich Felker 提交于
previously, a waiter could miss the 1->0 transition of block if another thread set block to 1 again after the signal function set block to 0. we now use the caller's thread id as a unique token to store in block, which no other thread will ever write there. this ensures that if block still contains the tid, no signal has occurred. spurious wakeups will of course occur whenever there is a spurious return from the futex wait and another thread has begun waiting on the cond var. this should be a rare occurrence except perhaps in the presence of interrupting signal handlers. signal/bcast operations have been improved by noting that they need not avoid inspecting the cond var's memory after changing the futex value. because the standard allows spurious wakeups, there is no way for an application to distinguish between a spurious wakeup just before another thread called signal/bcast, and the deliberate wakeup resulting from the signal/bcast call. thus the woken thread must assume that the signalling thread may still be waiting to act on the cond var, and therefore it cannot destroy/unmap the cond var.
-
由 Rich Felker 提交于
casting to int would not be correct because high bits could be lost. mapping the high bits down onto low bits would be costlier in the common case where the result is just used in a conditional. changing the type of the bit array elements to int would permute the order of the bit array on 64-bit big endian systems, so that's not an option either.
-
由 Rich Felker 提交于
-
- 23 9月, 2011 5 次提交
-
-
由 Rich Felker 提交于
actually this is just to avoid gcc being stupid and refusing to inline the function version, even when the size cost is essentially identical whether it's inlined or not.
-
由 Rich Felker 提交于
it's amazing none of the conformance tests i've run even bothered to check whether something so basic works...
-
由 Rich Felker 提交于
-
由 Rich Felker 提交于
-
由 Rich Felker 提交于
-
- 22 9月, 2011 10 次提交
-
-
由 Rich Felker 提交于
-
由 Rich Felker 提交于
-
由 Rich Felker 提交于
if the file descriptor resource limit has been increased past FD_SETSIZE, this is actually a security issue; we could write past the end of the fd_set object. using poll makes it a non-issue, and simplifies the code at the same time. also, use clock_gettime instead of gettimeofday, for reduced bloat and better entropy.
-
由 Rich Felker 提交于
-
由 Rich Felker 提交于
for now this is just a tiny optimization, but later if we support cancellation from __stdio_read and __stdio_write, it will be necessary for the recusrive lock count to be zero in order for these functions to know they are responsible for unlocking the FILE on cancellation.
-
由 Rich Felker 提交于
-
由 Rich Felker 提交于
-
由 Rich Felker 提交于
the arm syscall abi requires 64-bit arguments to be aligned on an even register boundary. these new macros facilitate meeting the abi requirement without imposing significant ugliness on the code.
-
由 Rich Felker 提交于
-
由 Rich Felker 提交于
-
- 21 9月, 2011 3 次提交
-
-
由 Rich Felker 提交于
-
由 Rich Felker 提交于
-
由 Rich Felker 提交于
-
- 20 9月, 2011 8 次提交
-
-
由 Rich Felker 提交于
-
由 Rich Felker 提交于
this is a case of poorly written man pages not matching the actual implementation, and why i hate implementing nonstandard interfaces with no actual documentation of how they're intended to work.
-
由 Rich Felker 提交于
this bug was introduced in a recent patch. the problem we're working around is that broken GNU software wants to use "struct siginfo" rather than "siginfo_t", but "siginfo" is not in the reserved namespace and thus not legal for the standard header to use.
-
由 Rich Felker 提交于
at the same time, make struct statfs match the traditional definition and make it more useful, especially the fsid_t stuff.
-
由 Rich Felker 提交于
-
由 Rich Felker 提交于
-
由 Rich Felker 提交于
really wchar_t should never vary, but the ARM EABI defines it as an unsigned 32-bit int instead of a signed one, and gcc follows this nonsense. thus, to give a conformant environment, we have to follow (otherwise L""[0] and L'\0' would be 0U rather than 0, but the application would be unaware due to a mismatched definition for WCHAR_MIN and WCHAR_MAX, and Bad Things could happen with respect to signed/unsigned comparisons, promotions, etc.). fortunately no rules are imposed by the C standard on the relationship between wchar_t and wint_t, and WEOF has type wint_t, so we can still make wint_t always-signed and use -1 for WEOF.
-
由 Rich Felker 提交于
this was the cause of crashes in printf when attempting to print floating point values.
-
- 19 9月, 2011 3 次提交
-
-
由 Rich Felker 提交于
-
由 Rich Felker 提交于
this port assumes eabi calling conventions, eabi linux syscall convention, and presence of the kernel helpers at 0xffff0f?0 needed for threads support. otherwise it makes very few assumptions, and the code should work even on armv4 without thumb support, as well as on systems with thumb interworking. the bits headers declare this a little endian system, but as far as i can tell the code should work equally well on big endian. some small details are probably broken; so far, testing has been limited to qemu/aboriginal linux.
-
由 Rich Felker 提交于
it does not work, but some configure scripts will falsely detect support then generate programs that crash when they call dlopen.
-