On Wed, May 13, 2015 at 04:23:42PM +0200, Aurelien Jarno wrote: > Alternatively as it triggers a segmentation fault, this could probably > be trapped and emulated in the kernel, just like it's done for the FPU > on some architectures.
The official erata list[1] tells us: | Problem: When a memory instruction with LOCK prefix executes and if it | encounters a page fault (#PF) the state of the CPU could potentially get | corrupted. There is no documented other outcome of such an operation, so can't be fixed up. The Intel SDK just stripped generating LOCK prefixes in the compiler. Bastian [1]: http://www.intel.com/content/dam/www/public/us/en/documents/specification-updates/quark-x1000-spec-update.pdf -- You're too beautiful to ignore. Too much woman. -- Kirk to Yeoman Rand, "The Enemy Within", stardate unknown -- To UNSUBSCRIBE, email to debian-glibc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/20150514150040.ga7...@mail.waldi.eu.org