On Tuesday 13 July 2010, Martyn Welch wrote: > > - lock_kernel(); > > + mutex_lock(&vme_user_mutex); > > ret = vme_user_ioctl(file->f_path.dentry->d_inode, file, cmd, arg); > > - unlock_kernel(); > > + mutex_unlock(&vme_user_mutex); > > > > Wha!, where'd that come from!!?? > > Ah, OK - "Staging: push down BKL into ioctl functions", missed that, > looks fine. > <http://git.kernel.org/gitweb.cgi?p=linux/kernel/git/sfr/linux-next.git;a=commitdiff;h=b1f2ac07636aadee5cb077fc7e830908b00fcaae> > Both patches only did blind conversions to drive the BKL out. Feel free to kill off that mutex and the vme_user_unlocked_ioctl() function in there again if you are convinced that no locking is actually needed in there. Arnd _______________________________________________ devel mailing list devel@xxxxxxxxxxxxxxxxxxxxxx http://driverdev.linuxdriverproject.org/mailman/listinfo/devel
