summaryrefslogtreecommitdiff
path: root/src/cpu/intel/hyperthreading
AgeCommit message (Collapse)Author
2014-05-10Replace SERIAL_CPU_INIT with PARALLEL_CPU_INITKyösti Mälkki
Lines with 'select SERIAL_CPU_INIT' where redundant with the default being yes. Since there is no 'unselect SERIAL_CPU_INIT' possibility, invert the default and rename option. This squelches Kconfig warnings about unmet dependencies. Change-Id: Iae546c56006278489ebae10f2daa627af48abe94 Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: http://review.coreboot.org/5700 Reviewed-by: Edward O'Callaghan <eocallaghan@alterapraxis.com> Tested-by: build bot (Jenkins) Reviewed-by: Patrick Georgi <patrick@georgi-clan.de> Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
2012-08-03Intel CPUs: Fix counting of CPU coresKyösti Mälkki
Detection for a hyper-threading CPU was not compatible with multicore CPUs. When using CPUID eax==4, also need to set ecx=0. CAR init tested on real hardware with hyper-threading model_f25 and under qemu 0.15.1 with multicore CPU. Change-Id: I28ac8790f94652e4ba8ff88fe7812c812f967608 Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: http://review.coreboot.org/1172 Tested-by: build bot (Jenkins) Reviewed-by: Anton Kochkov <anton.kochkov@gmail.com>
2012-07-31Revert "Use broadcast SIPI to startup siblings"Sven Schnelle
This reverts commit 042c1461fb777e583e5de48edf9326e47ee5595f. It turned out that sending IPIs via broadcast doesn't work on Sandybridge. We tried to come up with a solution, but didn't found any so far. So revert the code for now until we have a working solution. Change-Id: I7dd1cba5a4c1e4b0af366b20e8263b1f6f4b9714 Signed-off-by: Sven Schnelle <svens@stackframe.org> Reviewed-on: http://review.coreboot.org/1381 Tested-by: build bot (Jenkins) Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
2012-07-02Use broadcast SIPI to startup siblingsSven Schnelle
The current code for initializing AP cpus has several shortcomings: - it assumes APIC IDs are sequential - it uses only the BSP for determining the AP count, which is bad if there's more than one physical CPU, and CPUs are of different type Note that the new code call cpu->ops->init() in parallel, and therefore some CPU code needs to be changed to address that. One example are old Intel HT enabled CPUs which can't do microcode update in parallel. Change-Id: Ic48a1ebab6a7c52aa76765f497268af09fa38c25 Signed-off-by: Sven Schnelle <svens@stackframe.org> Reviewed-on: http://review.coreboot.org/1139 Tested-by: build bot (Jenkins) Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
2012-07-02Intel CPUs: execute microcode update only once per coreKyösti Mälkki
Early HT-enabled CPUs do not serialize microcode updates within a core. Solve this by running microcode updates on the thread with the smallest lapic ID of a core only. Also set MTRRs once per core only. Change-Id: I6a3cc9ecec2d8e0caed29605a9b19ec35a817620 Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: http://review.coreboot.org/1142 Tested-by: build bot (Jenkins) Reviewed-by: Sven Schnelle <svens@stackframe.org> Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
2012-05-08Clean up #ifsPatrick Georgi
Replace #if CONFIG_FOO==1 with #if CONFIG_FOO: find src -name \*.[ch] -exec sed -i "s,#if[[:space:]]*\(CONFIG_[A-Z0-9_]*\)[[:space:]]*==[[:space:]]*1[[:space:]]*\$,#if \1," {} + Replace #if (CONFIG_FOO==1) with #if CONFIG_FOO: find src -name \*.[ch] -exec sed -i "s,#if[[:space:]]*(\(CONFIG_[A-Z0-9_]*\)[[:space:]]*==[[:space:]]*1)[[:space:]]*\$,#if \1," {} + Replace #if CONFIG_FOO==0 with #if !CONFIG_FOO: find src -name \*.[ch] -exec sed -i "s,#if[[:space:]]*\(CONFIG_[A-Z0-9_]*\)[[:space:]]*==[[:space:]]*0[[:space:]]*\$,#if \!\1," {} + Replace #if (CONFIG_FOO==0) with #if !CONFIG_FOO: find src -name \*.[ch] -exec sed -i "s,#if[[:space:]]*(\(CONFIG_[A-Z0-9_]*\)[[:space:]]*==[[:space:]]*0)[[:space:]]*\$,#if \!\1," {} + (and some manual changes to fix false positives) Change-Id: Iac6ca7605a5f99885258cf1a9a2473a92de27c42 Signed-off-by: Patrick Georgi <patrick@georgi-clan.de> Reviewed-on: http://review.coreboot.org/1004 Tested-by: build bot (Jenkins) Reviewed-by: Ronald G. Minnich <rminnich@gmail.com> Reviewed-by: Martin Roth <martin@se-eng.com>
2012-03-25Intel cpus: Fix deadlock on hyper-threading initKyösti Mälkki
Only the BSP CPU was able to start its hyper-threading CPU siblings. When an AP CPU attempts this it calls start_cpu() within start_cpu(), deadlocking the system with start_cpu_lock. At the time intel_sibling_init() is run, the BSP CPU is still walking the cpu_bus linked list in lapic_cpu_init: start_other_cpus(). A sibling CPU appended at the end of this list will get started. Also fail compile with #error if SERIAL_CPU_INIT==0, as microcode updates on hyper-threading sibling CPUs must be serialized. Tested with HT-enabled P4 Xeons on dual-socket604 platform. Change-Id: I0053f58f49ed604605ce0a55e826d3e1afdc90b6 Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Reviewed-on: http://review.coreboot.org/775 Reviewed-by: Ronald G. Minnich <rminnich@gmail.com> Tested-by: build bot (Jenkins) Reviewed-by: Rudolf Marek <r.marek@assembler.cz>
2010-09-30Rename build system variables to be more intuitive, andPatrick Georgi
at the same time let the user specify sources instead of object files: - objs becomes ramstage-srcs - initobjs becomes romstage-srcs - driver becomes driver-srcs - smmobj becomes smm-srcs The user servicable parts are named accordingly: ramstage-y, romstage-y, driver-y, smm-y Also, the object file names are properly renamed now, using .ramstage.o, .romstage.o, .driver.o, .smm.o suffixes consistently. Remove stubbed out via/epia-m700 dsdt/ssdt files - they didn't easily fit in the build system and aren't useful anyway. Signed-off-by: Patrick Georgi <patrick.georgi@coresystems.de> Acked-by: Stefan Reinauer <stepan@coreystems.de> git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5886 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-04-27Since some people disapprove of white space cleanups mixed in regular commitsStefan Reinauer
while others dislike them being extra commits, let's clean them up once and for all for the existing code. If it's ugly, let it only be ugly once :-) Signed-off-by: Stefan Reinauer <stepan@coresystems.de> Acked-by: Stefan Reinauer <stepan@coresystems.de> git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5507 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-03-22printk_foo -> printk(BIOS_FOO, ...)Stefan Reinauer
Signed-off-by: Stefan Reinauer <stepan@coresystems.de> Acked-by: Ronald G. Minnich <rminnich@gmail.com> git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5266 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-02-07newconfig is no more.Patrick Georgi
Signed-off-by: Patrick Georgi <patrick.georgi@coresystems.de> Acked-by: Ronald G. Minnich <rminnich@gmail.com> git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5089 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2009-08-12Kconfig!Patrick Georgi
Works on Kontron, qemu, and serengeti. Signed-off-by: Patrick Georgi <patrick.georgi@coresystems.de> tested on abuild only. Acked-by: Ronald G. Minnich <rminnich@gmail.com> git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4534 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2009-06-03Revert "CMOS: Add set_option and rework get_option."Luc Verhaegen
This reverts commit eb7bb49eb5b48c39baf7a256b7c74e23e3da5660. Stepan pointed out that "s" means string, which makes the following statement in this commit message invalid: "Since we either have reserved space (which we shouldn't do anything with in these two functions), an enum or a hexadecimal value, unsigned int seemed like the way to go." Signed-off-by: Luc Verhaegen <libv@skynet.be> Acked-by: Luc Verhaegen <libv@skynet.be> git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4335 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2009-06-03CMOS: Add set_option and rework get_option.Luc Verhaegen
To ease some of my debugging pain on the unichrome, i decided i needed to move FB size selection into cmos, so i could test a size and then reset it to the default after loading this value so that the next reboot uses the (working) default again. This meant implementing set_option in parallel to get_option. get_option was then found to have inversed argument ordering (like outb) and passing char * and then depending on the cmos layout length, which made me feel quite uncomfortable. Since we either have reserved space (which we shouldn't do anything with in these two functions), an enum or a hexadecimal value, unsigned int seemed like the way to go. So all users of get_option now have their arguments inversed and switched from using ints to unsigned ints now. The way get_cmos_value was implemented forced us to not overlap byte and to have multibyte values be byte aligned. This logic is now adapted to do a full uint32_t read (when needed) at any offset and any length up to 32, and the shifting all happens inside an uint32_t as well. set_cmos_value was implemented similarly. Both routines have been extensively tested in a quick separate little program as it is not easy to get this stuff right. build_opt_tbl.c was altered to function correctly within these new parameters. The enum value retrieval has been changed strol(..., NULL, 10) to stroul(..., NULL, 0), so that we not only are able to use unsigned ints now but so that we also interprete hex values correctly. The 32bit limit gets imposed on all entries not marked reserved, an unused "user_data" field that appeared in a lot of cmos.layouts has been changed to reserved as well. Signed-off-by: Luc Verhaegen <libv@skynet.be> Acked-by: Peter Stuge <peter@stuge.se> git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4332 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2009-04-22Trivial removal of a freudian slip.Patrick Georgi
Signed-off-by: Patrick Georgi <patrick.georgi@coresystems.de> Acked-by: Patrick Georgi <patrick.georgi@coresystems.de> git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4164 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2009-02-28coreboot-v2: drop this ugly historic union name in v2 that was dropped in v3Stefan Reinauer
a long time ago. This will make it easier to port v2 boards forward to v3 at some point (and other things) Signed-off-by: Stefan Reinauer <stepan@coresystems.de> Acked-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006@gmx.net> git-svn-id: svn://svn.coreboot.org/coreboot/trunk@3964 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2008-08-01drop unused code (trivial)Stefan Reinauer
Signed-off-by: Stefan Reinauer <stepan@coresystems.de> Acked-by: Stefan Reinauer <stepan@coresystems.de> git-svn-id: svn://svn.coreboot.org/coreboot/trunk@3452 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2005-09-14Print a failure message if a sibling CPU fails to start.Steven J. Magnani
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@2031 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2004-10-14- Add new cvs code to cvsEric Biederman
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@1657 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1