You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
At https://github.com/jnovy/pxz/blob/master/pxz.c#L338, the allocated memory is not freed. The problem is that when there are many files to be processed at once (e.g., ./pxz -k -T 8 -q -f c-input/*, and since the maximum argument length is 131072 on my machine, c.f. here), it may eat up memories -- the allocations are inside a loop traversing the input files. Since each time the leaked memory is 25169920 on my machine, 25169920*131072=3.2990718e+12bytes, which is roughly 3200Gigabytes.
When the command lasts for a long time, it may also slowdown the host machine.
Also, as seen in #31, there is no upper bound limit for the specified thread number, this may lead to quick segfault.
The text was updated successfully, but these errors were encountered:
At https://github.com/jnovy/pxz/blob/master/pxz.c#L338, the allocated memory is not freed. The problem is that when there are many files to be processed at once (e.g.,
./pxz -k -T 8 -q -f c-input/*
, and since the maximum argument length is 131072 on my machine, c.f. here), it may eat up memories -- the allocations are inside a loop traversing the input files. Since each time the leaked memory is 25169920 on my machine, 25169920*131072=3.2990718e+12bytes, which is roughly 3200Gigabytes.When the command lasts for a long time, it may also slowdown the host machine.
Also, as seen in #31, there is no upper bound limit for the specified thread number, this may lead to quick segfault.
The text was updated successfully, but these errors were encountered: