forwarded 712387 https://bugs.webkit.org/show_bug.cgi?id=42756 tag 712387 upstream thanks On Saturday 15 June 2013 17:33:43 Török Edwin wrote: [snip very nice debugging info] > See here for some discussions upstream, but leading to no solution (still > allocates 2GB): https://bugs.webkit.org/show_bug.cgi?id=42756 > > It seems silly to allocate 2GB just to avoid having jumps crossing 2GB > boundaries, when there is a dedicated mmap() flag that ensures all > addresses are within 1st 2GB: MAP_32BIT. > > I'd suggest to use that flag and set the VM_POOL_SIZE to 32MB like it is for > 32-bit apps. While at first sight I agree with you, this has to be decided by upstream. We do not patch things without upstream's ACK *except* on very specific cases (security bugs, obvious fixes, etc.). This change could lead to some other undesired behaviour, so please ask upstream to fix it (normally submitting patches to upstream is a good way to help). > It would be best if the JIT could also be disabled at runtime for > libqt4-script, for applications that don't strictly require it, like > plasma, etc. This is the case for another bug, severity wishlist for *upstream* (in this case, Qt). Please do not heasitate in filling it in https://bugreports.qt-project.org/ Kinds regards, Lisandro. -- Lisandro Damián Nicanor Pérez Meyer http://perezmeyer.com.ar/ http://perezmeyer.blogspot.com/
Attachment:
signature.asc
Description: This is a digitally signed message part.