| Age | Commit message (Collapse) | Author |
|---|
|
Closing files in these callbacks could block scheduler progress
and cause major system instability. We now defer these operations
to a dedicated process instead.
This process may in turn block forever and prevent further orphaned
files from being closed, but it will keep the emulator itself from
misbehaving.
|
|
|
|
|
|
It wasn't possible to change group/owner separately, and our test
suite lacked coverage for that.
ERL-589
|
|
To the best of our knowledge this was introduced since file operations
on device files/FIFO:s could hang the emulator forever back when the
emulator was single-threaded and lacked IO threads; a read operation
could block all progress preventing the write operation it waited for
from occurring.
Granted, this could still happen through starving all dirty IO
schedulers, but the same issue can arise with NFS files which we've
always allowed.
Removing this restriction also lets us remove a stat(2) call that was
added to specifically allow `/dev/null`.
|
|
This improves the latency of file operations as dirty schedulers
are a bit more eager to run jobs than async threads, and use a
single global queue rather than per-thread queues, eliminating the
risk of a job stalling behind a long-running job on the same thread
while other async threads sit idle.
There's no such thing as a free lunch though; the lowered latency
comes at the cost of increased busy-waiting which may have an
adverse effect on some applications. This behavior can be tweaked
with the +sbwt flag, but unfortunately it affects all types of
schedulers and not just dirty ones. We plan to add type-specific
flags at a later stage.
sendfile has been moved to inet_drv to lessen the effect of a nasty
race; the cooperation between inet_drv and efile has never been
airtight and the socket dying at the wrong time (Regardless of
reason) could result in fd aliasing. Moving it to the inet driver
makes it impossible to trigger this by closing the socket in the
middle of a sendfile operation, while still allowing it to be
aborted -- something that can't be done if it stays in the file
driver.
The race still occurs if the controlling process dies in the short
window between dispatching the sendfile operation and the dup(2)
call in the driver, but it's much less likely to happen now.
A proper fix is in the works.
--
Notable functional differences:
* The use_threads option for file:sendfile/5 no longer has any
effect.
* The file-specific DTrace probes have been removed. The same
effect can be achieved with normal tracing together with the
nif__entry/nif__return probes to track scheduling.
--
OTP-14256
|
