Page MenuHomeFreeBSD

riscv trap_pfault: remove unneeded hold of the process around vm_fault() call.
ClosedPublic

Authored by kib on Sep 13 2019, 7:22 PM.
Tags
None
Referenced Files
Unknown Object (File)
Mon, Sep 21, 4:05 PM
Unknown Object (File)
Sun, Sep 20, 12:43 PM
Unknown Object (File)
Fri, Sep 18, 8:58 PM
Unknown Object (File)
Fri, Sep 18, 8:16 PM
Unknown Object (File)
Fri, Sep 18, 6:11 PM
Unknown Object (File)
Thu, Sep 17, 12:34 PM
Unknown Object (File)
Sun, Sep 13, 5:52 PM
Unknown Object (File)
Sun, Sep 13, 1:30 PM
Subscribers

Details

Summary

This is re-appearance of the nop code removed from other arches in r287625.

Diff Detail

Repository
rS FreeBSD src repository - subversion
Lint
Lint Not Applicable
Unit
Tests Not Applicable

Event Timeline

markj added inline comments.
sys/riscv/riscv/trap.c
240 ↗(On Diff #62059)

Wasn't this incorrect for faults on pipe_map or exec_map?

This revision is now accepted and ready to land.Sep 13 2019, 7:42 PM
sys/riscv/riscv/trap.c
240 ↗(On Diff #62059)

Incorrect what ? pipe or exec maps are kernel submaps. What do you mean exactly ?

sys/riscv/riscv/trap.c
240 ↗(On Diff #62059)

Sorry, I meant that the comment seemed incorrect. I don't see why we would lock a process for faults on a user map, but not a fault on pipe_map. Anyway, it is not important.