Date: Sun, 26 Sep 1999 19:22:47 +0200 (CEST) From: Ingo Molnar <mingo@chiara.csoma.elte.hu> To: linux-smp@vger.rutgers.edu, linux-kernel@vger.rutgers.edu Subject: [x86,SMP,patch] smp-2.3.18-H3 the latest version of the x86 SMP/APIC/IOAPIC code can be found at: http://www.redhat.com/~mingo/smp-2.3.18-H3 this one is a 'release candidate', ie. i fixed all reported problems - everybody please check wether there are unfixed SMP problems left (or new ones introduced) and re-report them if so. (just in case i missed something) Changes in smp-2.3.18-H3: - finally fixes Mike Frisch's 'soundblaster IRQ goes berzerk' problem. (me) Changes in smp-2.3.18-H2: - we should boot on boxes with certain default mptables now. (Ken-ichi Yaku, Alvaro Herrera) - NMI handling fixes, typo fixes (Maciej W. Rozycki) - fixed minor bug in smp_call_function(), cleaned it up, converted to semaphore usage. Other cleanups. (me) Changes in smp-2.3.18-F8: - continuing quest to make 82489DX based (and other older) systems work, typo fixes and comment updates. (Maciej W. Rozycki) - 'local APIC timer interrupt interleaving', this is both a fix for a potential bug, and preparation for next generation per-CPU timer handling. (me) - tons of further cleanups, separated smp.c into smpboot.c and smp.c, cleaned up both. (me) - potential double-write bug fixed - this might fix the booting problem on those old ALR dual- and quad-Pentium boards? (if anyone got one) (me) - other smaller fixes i forgot about Changes in smp-2.3.18-B1: - unconditional NMI oopser. It turned out that we can do this on 99% (maybe 100%) of x86 SMP boxes without impacting timer IRQ performance. See Documentation/nmi_watchdog.txt. (me) - old external IOAPICs and 486-based SMP boxes should work now (Maciej W. Rozycki) - unused variable space cleanups, initialize_secondary() cleanup (Manfred Spraul) - moved all IO-APIC and IPI APIC messages to logical destination mode, fixed LDR initialization bug, cleaned up irq-vector space. (me) - 'timer-sync' feature, all CPUs synchronize their time stamp counters, errors get reported and fixed. This fixes certain dual-PIII boards. (me) - the 'set the destination CPU mask to 0' IOAPIC-edge trick is actually not valid and triggers APIC bugs. Exchanged it for an equivalent but safe method. (me) - 8259A init moved out of setup.S (Maciej W. Rozycki) - spurious 8259A interrupts get reported properly, APIC error vector exported and logged properly (me) - a 'noapic' bug fixed - fixes instability in noapic mode. (me) - we can inline ack_APIC() on P6 boxes: smaller and faster IRQ code. (me) - other stuff. reports, comments, suggestions welcome. -- mingo - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/