If multiple threads were waiting for more block buffers without anyone
releasing them, they ended up in a deadlock.
Now we store 6 blocks for 8 threads. If a thread already has a block
buffer, it will not have to wait for a new one. Only if there are more
than 8 threads using blocks, will it block until there are free slots
for a thread available.
Add support for processor local futexes. These work the exact same way
as global ones, but only lock a process specific lock and use a process
specific hash map.
Also reduce the time futex lock is held. There was no need to hold the
global lock while validating addresses in the process' address space.
If the processor has invariant TSC it can be used to measure time. We
keep track of the last nanosecond and TSC values and offset them based
on the current TSC. This allows getting current time in userspace.
The implementation maps a single RO page to every processes' address
space. The page contains the TSC info which gets updated every 100 ms.
If the processor does not have invariant TSC, this page will not
indicate the capability for TSC based timing.
There was the problem about how does a processor know which cpu it is
running without doing syscall. TSC counters may or may not be
synchronized between cores, so we need a separate TSC info for each
processor. I ended up adding sequence of bytes 0..255 at the start of
the shared page. When a scheduler gets a new thread, it updates the
threads gs/fs segment to point to the byte corresponding to the current
cpu.
This TSC based timing is also used in kernel. With 64 bit HPET this
probably does not bring much of a benefit, but on PIT or 32 bit HPET
this removes the need to aquire a spinlock to get the current time.
This change does force the userspace to not use gs/fs themselves and
they are both now reserved. Other one is used for TLS (this can be
technically used if user does not call libc code) and the other for
the current processor index (cannot be used as kernel unconditionally
resets it after each load balance).
I was looking at how many times timer's current time was polled
(userspace and kernel combined). When idling in window manager, it was
around 8k times/s. When running doom it peaked at over 1 million times
per second when loading and settled at ~30k times/s.
Joystick axis and buttons are now named to standard values, this allows
interfacing multiple different controllers (only DS3 is supported)
Add ioctl calls for userspace to set joystick player leds and rumble
Only use DS3 code paths when we detect that the attached device is
actually an DS3 controller
update test-joystick program to the new interface and add support to
control rumble and player leds
I was creating a local variable shadowing the global one. This prevented
cleanup to close it. (this is not really necessary as the program dies
anyway)
I'm not sure if these are used by anything but I would assume so as I
have added them :D
functions added:
- getprotobyname
- open_memstream
- munlock
- lockf
- nice
- crypt
- getsid
- wcstoul
At the moment these lock the whole file which is not what is supposed to
happen. Some port was trying to use these and this seems to work for
that. This may cause deadlocks but that should be easy enough to find
the reason because of the debug warnings
Before this real hardware failed to boot with smp enabled. Allocating
the idle thread does a page mapping which ends up broadcasting TLB
shootdown to other processes. This ends up failing somewhere halting the
processors never allowing them to initialize their scheduler
Each allocated inode used to call sync(). Each sync reads and writes
a block from the filesystem. Doing a 1 MiB write ended up syncing around
257 times
Small symlinks store their target in inline storage of data block
indices. Freeing blocks did not account for this and tried to interpret
the link target as block inidices to delete
Bananymous