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
this is a known bug in DR but hasn't been fixed since hasn't cause problems yet, but filing in issue database since now I see it on python:
Linux RT signals actually include 64 as a valid signal number.
The Linux kernel subtracts one from signal numbers before using as
array or bitmask refs, to keep its structs an even 64 size.
This is very confusing and various info is contradictory,
but it seems that there are 33 RT signals, numbered 32..64.
running 64-bit python on my FC12 box hits this issue
the complication is w/ 0-based vs 1-based signal numbers (b/c 0 unused) which is done for sigsets but not elsewhere so the fix requires more than just bumping the constant (else I would have done that instead of adding the comment): needs two different constants or changes to the loops or sthg.
From [email protected] on August 11, 2010 13:06:03
this is a known bug in DR but hasn't been fixed since hasn't cause problems yet, but filing in issue database since now I see it on python:
Linux RT signals actually include 64 as a valid signal number.
The Linux kernel subtracts one from signal numbers before using as
array or bitmask refs, to keep its structs an even 64 size.
This is very confusing and various info is contradictory,
but it seems that there are 33 RT signals, numbered 32..64.
running 64-bit python on my FC12 box hits this issue
<Application python (5887). Internal Error Internal DynamoRIO Error: /work/dr/tot/opensource/core/linux/signal.c:1686 sig < MAX_SIGNUM
(Error occurred @5975 frags)
Original issue: http://code.google.com/p/dynamorio/issues/detail?id=336
The text was updated successfully, but these errors were encountered: