Demi Marie Obenour
Demi Marie Obenour
LinuxThreads hasn’t been in use for a long time, replaced by the Native POSIX Thread Library.
Can you build the runtime with undefined behavior sanitizer in a mode where it detects integer overflows, underflows, wraparound, and truncation?
@pottering: pidfds can already be used for killing process without needing to also use them to obtain their exit status. `waitid(P_PID)` is already race-free. The best algorithm I could find...
@poettering ping :smile: Thanks to `pidfd_open` and friends, it is now possible to wait for a child process to exit without any races. It is also possible to get a...
@poettering: The problem with `waitid(P_ALL)` can be worked around by using `WNOWAIT` and then reaping only the expected children.
> > > ["misbehaving (...) application modifies a buffer in-flight when doing an O_DIRECT write"](https://lore.kernel.org/linux-btrfs/c3cc0815c5756d07201c57063f3759250f662c77.1600961206.git.josef@toxicpanda.com/). In our case the application is xen-blkback on a loop device > > > it...
@rustybird I can have the not-script check the filesystem type (via `fstatfs(2)`) and enable direct I/O only on ext4 or XFS (where it is a free performance win). There might...
Modifying a buffer that is in-flight is a straightforward data race and undefined behavior. No application should ever do this. The only difference here is that when the junk data...
You will need to ensure that the relevant kernel driver (such as i915) is not loaded in dom0. GPU drivers generally cannot be unloaded or unbound from the respective devices.
@preland What happens if you boot with `nomodeset` on the Linux kernel command line? AMD does not support PCI passthrough for their client GPUs so I am not surprised that...