$Id: RELEASE-NOTES,v 1.234.2.166 2010/11/07 19:48:14 mikpe Exp $

RELEASE NOTES

=============

Version 2.6.42, 2010-11-07

- x86.c: identify Intel Family 6 Models 37 and 44 as Westmere not Nehalem.

- x86.c: do_init_tests() calls perfctr_x86_init_tests() which is __init,

  therefore mark do_init_tests() also as __init.

Version 2.6.41, 2010-06-08

- x86: Add support for OFFCORE_RSP_{0,1} on Nehalem/Westmere.

- x86: Recognise Intel family 6 models 30 and 37 as Nehalems.

  Update comments mapping product lines to model numbers.

- x86: Rename PERFCTR_X86_INTEL_COREI7 CPU/PMU type constant to

  PERFCTR_X86_INTEL_NHLM. Update driver to print "Nehalem" rather

  than "Core i7" when a CPU of this type is detected.

Version 2.6.40, 2010-01-30

- x86: add comment after #endif terminating big kernel >= 2.6.19 block

- x86: handle cpumask API change in kernel 2.6.32

- x86: recognize Intel Family 6 Model 2Eh processors (Nehalem Xeon 7500).

- x86: recognize Intel Family 6 Model 2Ch processors (i7-980X, Gulftown).

- x86: recognize AMD Family 11h processors, support them as 10h ones.

Version 2.6.39, 2009-06-11

- ppc and arm: updates to match perfctr_cpu_update_control() changes,

  add missing #include <asm/cputype.h> to arm.c

- global.c: coding style fixups

- x86.c: update AMD multicore detection to match the documentation

  and actually work on current processors, set up cpumask of all

  core0 CPUs, detect RevE processors, update p6_like_check_control()

  to allow per-thread sessions to use AMD NB events on post-RevE

  processors but limit them to core0 CPUs

- x86.c: replace is_global parameter to perfctr_cpu_update_control()

  with a cpumask_t pointer, make P4 update this cpumask instead of

  hard-coding the use of perfctr_cpus_forbidden_mask in virtual.c,

  add cpumask to struct vperfctr, update virtual.c to use the cpumask

  from perfctr_cpu_update_control() not perfctr_cpus_forbidden_mask

  to derive the task's new cpumask, update set_cpus_allowed() callback

  to validate new cpumask against the vperfctr's private one, update

  global.c to pass a NULL cpumask_t pointer

- virtual.c: rearrange sys_vperfctr_control() so that set_cpus_allowed()

  comes after perfctr_cpu_update_control(), record updater's ->tgid so

  races with concurrent updaters can be detected and handled

- virtual.c: make vperfctrfs_dentry_operations 'const' in 2.6.30 and

  later kernels.

- x86.c: silence MSR_IA32_MISC_ENABLE_PEBS_UNAVAIL redefinition warning.

Version 2.6.38, 2009-01-23

- Remove 2.4 kernel support from <linux/perfctr.h> (cpumask_t workaround).

- Remove 2.4 kernel support from build system.

- Remove 2.4 kernel support from .c files.

- Remove 2.4 kernel support from .h files.

- Kernel 2.6.29-rc1 changed remap_pfn_range() to WARN_ON when applied

  to plain RAM. Update virtual.c to use vm_insert_page() instead.

- Kernel 2.6.29-rc1 moved a task's fsuid/fsgid field to the ->cred

  struct. Update virtual.c to use current_fsuid() and current_fsgid().

  Update compat.h to supply these macros for older kernels.

Version 2.6.37, 2008-11-30

- x86: Preliminary Intel Core i7 support, limited to handling it

  as a Core2-like processor with four PMCs. The AnyThread evntsel

  flag and off-core/uncore monitoring are not yet supported.

- x86: Recognise Intel Family 6 Model 29 (Xeon 7400) as Core 2.

- x86: Make core2_clear_counters() also clear the FREEZE_PERFMON_ON_PMI

  bit in DEBUGCTLMSR. vtune leaves this bit set, which breaks perfctr.

- x86: Make perfctr_clear_counters() initialise MSR_CORE_PERF_GLOBAL_CTRL

  during initialisation. This fixes compatibility issues with drivers

  that may leave this register cleared == all counters disabled. vtune

  is reported to do this. Add ->clear_counters() op to perfctr_pmu_msrs

  to handle this cleanly. Convert Via C3 to use this mechanism instead

  of being a special case.

Version 2.6.36, 2008-10-19

- x86: Limit the value written to a fixed-function counter's MSR

  to 40 bits. Extraneous high bits cause GP faults on Model 23

  Core2s, while earlier processors would just ignore them.

- Kernel 2.6.27-rc1 dropped the retry parameter to on_each_cpu()

  and smp_call_function(). Adjust accordingly. Add compatibility

  wrappers for older kernels.

- Kernel 2.6.27-rc1 removed find_task_by_pid(). Migrate to new

  find_task_by_vpid(). Add compatibility wrapper for older kernels.

- Starting with kernel 2.6.27-rc1 one should use alloc_intr_gate()

  not set_intr_gate() on x86 when binding a specific vector, as this

  also marks the vector as allocated. Adjust <asm-x86/perfctr.h>.

Version 2.6.35, 2008-06-30

- x86: Preliminary Intel Atom support:

  * add Atom CPU type, it differs from all previous models

  * Atom is poorly documented, so query cpuid leaf 0xA for

    its architectural PMU capabilities; initial Atoms appear

    to have APM V3, 2 40-bit general-purpose counters, 7

    architectural events, and 1 40-bit fixed-function counter

- x86: Replace the p6_is_core2 flag with separate variables

  indicating (a) having per-evntsel enable bits, and (b) the

  number of fixed-function counters available.

- x86: intel_p6_init(): recognise Celeron model 16h and

  treat it as a Core 2.

- x86: Clean up intel_p6_init(): replace complex if conditions

  with explicit switches on x86_model, explicitly enumerate

  accepted model numbers.

- x86: Correct p6_like_check_control() to reject regular

  pmcs >= 2 on Core2 before mapping fixed-function counters

  0x40000000+N to pmcs 2+N for duplicate counter checking.

  The failure to reject those invalid pmcs made it possible

  for users to cause the driver to perform invalid wrmsr and

  rdpmc accesses with kernel hangs as the result.

Version 2.6.35-pre1, 2008-06-23

- Add optional close-on-exec feature for per-process perfctrs:

  * reassign _reserved1 as flags in vperfctr_control

  * add VPERFCTR_CONTROL_CLOEXEC flag

  * add perfctr_flush_thread() hook to exec() path

  * map perfctr_flush_thread() to __vperfctr_flush() via

    inline functions and virtual_stub.c

  * in __virtual_flush(), if CLOEXEC is set then unlink the state

  * bump API version to 5.2

Version 2.6.34, 2008-05-29

- Reorder kernel version and HAVE_EXPORT___put_task_struct

  tests in compat.h to handle the SuSE 2.6.16.42-0.12 kernel

  exporting __put_task_struct_cb().

- Fix warning about DONT_HAVE_i_blksize being undefined in

  the SuSE 2.6.16.42-0.12 kernel.

Version 2.6.33, 2008-05-18

- x86: Intel Family 6 Model 23 support missed that it needs

  to trigger LVTPC reinit. Fix that.

Version 2.6.32, 2008-04-20

- x86: Recognize Intel Family 6 Model 23 as Core2.

- x86: Update perfctr_sysclass definition for kernel 2.6.25.

Version 2.6.31, 2008-01-26

- x86: Correct Barcelona CPU type to read FAM10H not FAM10

  w/o the trailing H. In struct perfctr_cpu_control, place

  p4 struct in a union and alias p4.escr[] with envtsel_high[]:

  this allows passing high 32 evntsel bits for Barcelona.

  Update driver to also manage high 32 evntsel bits on Barcelona,

  on other processors those bits are forced to zero.

Version 2.6.30, 2007-10-28

- x86: Kernel 2.6.24-rc1 changed the calling convention for

  the cpu_data macro. Updated accordingly, and added compat

  code providing the new behaviour in older kernels.

- ppc32: Kernel 2.6.24-rc1 removed the get_property() compat

  macro. Use of_get_property() with kernels >= 2.6.22.

- The workaround for RHEL5 removing ptrace_check_attach()

  only works when perfctr is built as a module. Fix it to

  also work in the non-modular case.

Version 2.6.29, 2007-10-07

- Add new cpu_type for AMD Family 10h, to reduce confusion.

- Very preliminary support for AMD Family 10h processors.

  They will need a new cpu_type and support for 64-bit evntsels.

  For now pretend they are K8C processors.

- Silence compilation warnings on ppc32.

- Intel has finally documented how to read the Core 2's

  fixed-function performance counters in user-space:

  rdpmc 0x4000000N for N=0,1,2. Support them from user-space

  by pretending they have P6-like evntsels, and extract the

  useful controls (CPL+INT) into the fixed-function counters

  control register. Update the P6 driver methods to handle the

  fact that a fixed-function counter has no private evntsel MSR.

  Update x86_tests to measure the cost of reading these counters

  and writing their shared control register.

- More Intel CPU detection cleanup: separate detection needed

  by the driver from that done to supply cpu_type to user-space.

- Clean up Intel CPU detection by moving family 5, 6, and 15

  detection code to separate procedures.

- Use #undef to silence macro redefinition warnings on x86.

- Kernel 2.6.22 removed the rdtsc() macro from i386. Unbreak

  and clean up x86_tests.c by using our own rdtsc_low() macro.

Version 2.6.28, 2007-07-18

- x86: The kernel's perfctr/nmi management system changed again

  in 2.6.22-rc5. Call {disable,enable}_lapic_nmi_watchdog() not

  {stop,setup}_apic_nmi_watchdog() in kernels >= 2.6.22.

- ppc.c: kernel 2.6.22-rc1 removed find_type_devices(),

  use of_find_node_by_type() and of_node_put() instead

- compat.h: fix warnings in CONFIG_UTRACE check

Version 2.6.27, 2007-04-09

- Bumped copyright years on recently updated files.

- RHEL5 2.6.18-8.1.1.el5 added export of __put_task_struct.

  Patched <linux/config.h> to signal this.

- RHEL5 2.6.18-8.1.1.el5 removed i_blksize. Patched

  <linux/config.h> to signal this. Check this in virtual.c.

- RHEL5 2.6.18-8.1.1.el5 replaced ptrace with utrace, breaking

  the remote control API which needs ptrace_check_attach().

  For now, stub ptrace_check_attach() so that things build.

- The {reserve,release}_{perfctr,evntsel}_nmi() API changed

  in kernel 2.6.21-rc6, from being CPU-local to being global.

  Updated x86.c to handle this change.

Version 2.6.26, 2007-02-11

- Updates to show my @it.uu.se email address in some messages,

  as the old @csd.uu.se address now is /dev/null.

- Added driver support for ARM/XScale processors. Overflow

  interrupts are not yet supported, in part due to conflicts

  with Intel's ixp400_eth driver. Plain event counting works.

- Kernel 2.6.20-rc1 moved filp->f_dentry and filp->f_vfsmnt into

  the filp->fpath substructure. Added compat macros to handle this.

- Kernel 2.6.20-rc1 changed how pipefs handles its dentries.

  Adapted those changes to vperfctrfs.

Version 2.6.25, 2006-10-15

- x86.c: Intel Core 2 is substantially different from Intel Core.

  Add new cpu_type for Core 2, map family 6 model 15 to Core 2,

  and require Core 2 to set Enable in all EVNTSELs.

- x86.c: kernel 2.6.19-rc1 removed the {reserve,release}_lapic_nmi()

  API, and added a {reserve,release}_{perfctr,evntsel}_nmi() API.

  Reimplement {reserve,release}_lapic_nmi(). Add data to describe

  the set of perfctr, evntsel, and other MSRs used by a CPU type.

  Add procedures to reserve and release all of the MSRs. Replace

  the CPU-specific clear_counters() procedures with a generic one

  that uses the MSR description data object. Add EXPORT_SYMBOL of

  {setup,stop}_apic_nmi_watchdog() to x86_setup.c for 2.6.19+ kernels.

- virtual.c: kernel 2.6.18 dropped EXPORT_SYMBOL(tasklist_lock),

  so starting with 2.6.18 we must use rcu_read_{lock,unlock}()

  around find_task_by_pid().

- virtual.c: kernel 2.6.19-rc1 dropped the inode->i_blksize field.

- Only #include <linux/config.h> for kernels older than 2.6.19,

  since 2.6.19-rc1 marks it deprecated. The test is ugly: perhaps

  this should be handled in the Makefile instead.

Version 2.6.24, 2006-09-17

- x86_tests: fixed linkage error caused by p6_init_tests()

  not being compiled in 64-bit builds.

Version 2.6.23, 2006-08-20

- Testing done by the PAPI folks indicate that Intel Core 2 has

  a single master Enable bit in EVNTSEL0, just like previous P6s.

  Restore Intel Core to the classic P6 rule: EVNTSEL0 must be

  enabled, EVNTSEL1 must not be enabled.

- Intel Core updates: each EVNTSEL has its own Enable bit like

  AMD and P4, recognise Model 15 (Core2), Core2 is 64-bit so

  make P6 testing code available in both 32- and 64-bit builds.

- ppc32: correct PMC1SEL and PMC4SEL definitions.

- virtual: new vperfctrfs_get_sb() for kernel 2.6.18-rc1 and

  later: ->get_sb() and get_sb_pseudo() changed prototype.

- x86: #include <asm/nmi.h> to get lapic NMI declarations in

  kernel 2.6.18-rc1 and later. Do not do this if we're going

  to stub them because !CONFIG_X86_LOCAL_APIC.

Version 2.6.22, 2006-06-02

- Preliminary support for Intel Core (family 6 model 14) processors.

- x86: The code to extract max_cores_per_package from CPUID(4):EAX

  needs cpuid() to put zero in ecx, but it only does that in fairly

  new 32-bit kernels, not in 64-bit kernels or older 32-bit kernels.

  This badly broke the SMT_ID detection on a dual-processor dual-core

  hyper-threaded 64-bit Xeon machine. Fixed by using cpuid_count()

  instead. Added compatibility definition of it for kernels < 2.6.12.

- Fixed x86_tests.c compilation error in the i386 2.6.16 kernel

  by moving sync_core() definition from x86_tests.c to x86_compat.h

  and only defining it in i386 kernels older than 2.6.16.

Version 2.6.21, 2006-04-03

- Converted mutex-like semaphores to the new mutex type

  introduced in kernel 2.6.16. Added simulation of the

  new API in terms of semaphores to compat.h and compat24.h.

- put_task_struct() uses __put_task_struct() again starting

  with the 2.6.17-rc1 kernel. Updated compat.h, compat24.h,

  and virtual_stub.c accordingly.

- Corrected a botched cleanup of compat24.h in perfctr-2.6.20

  which broke support for RHEL3 2.4.21 kernels.

Version 2.6.20, 2006-03-12

- Starting with 2.6.16-rc1, put_task_struct() uses an RCU callback

  __put_task_struct_cb() instead of the old __put_task_struct().

  2.6.16-rc6 dropped the EXPORT_SYMBOL() of __put_task_struct_cb().

  Updated compat.h, compat24.h, and virtual_stub.c accordingly.

Version 2.6.19, 2006-01-22

- Updated ppc32 driver for kernel 2.6.16-rc1: dynamically

  claim the HW and register our interrupt handler via

  {reserve,release}_pmc_hardware(); simulate these primitives

  in older kernels; fully migrate patch kit from arch/ppc/

  to arch/powerpc/.

Version 2.6.18, 2006-01-03

- 2.6.5-7.201-suse added EXPORT_SYMBOL_GPL(__put_task_struct).

  Added feature #define to the kernel patch, and modified compat.h

  to disable our export of __put_task_struct in this case.

- Merged the structure descriptor declarations in marshal.c

  to avoid duplicating the parts that are identical across

  all supported platforms.

Version 2.6.17, 2005-10-02

- The dual-core P4s changed the layout rules for the initial

  APIC ID, which broke the x86 driver on DC P4s. Updated the

  HT thread ID detection code to match current IA32 SDM Vol3.

- Kernel 2.4.21-37.EL added EXPORT_SYMBOL_GPL(__put_task_struct).

  Added new feature #define to the kernel-specific patch for this

  case. compat24.h now disables our export of __put_task_struct

  when that feature #define is set.

- Kernel 2.6.14-rc1 changed the state parameter to ->suspend()

  methods to be of type 'pm_message_t'. Adjusted x86.c for this,

  to eliminate a compile-time type mismatch warning.

Version 2.6.16, 2005-09-04

- cpu_khz changed type in kernel 2.6.13. Adjusted x86_setup.c

  accordingly, to avoid a compile-time error.

- The ppc32 driver will now compile in kernels that lack Open

  Firmware support, which is needed for some embedded systems.

Version 2.6.15, 2005-05-06

- Added code to detect multicore K8s and prevent threads in the

  thread-centric API from using northbridge events. This avoids

  resource conflicts, and an erratum in Revision E chips.

- #undef MMCR0_PMXE in ppc_compat.h, to avoid macro redefinition

  complaints in 2.6 kernels.

Version 2.6.14, 2005-04-09

- x86: Reverted the workaround in perfctr-2.6.13 for the problem

  that gcc-4.0 snapshots appeared to ignore 'noinline' on static

  functions, as recent gcc-4.0 prereleases seem to work correctly.

Version 2.6.13, 2005-02-13

- global.c: Allow user-space to disable the in-kernel sampling

  timer by setting interval_usec == 0 in the START command.

  In this case sampling is done by the READ command.

- Modified x86 call backpatching code to avoid breaking with

  gcc-4.0 snapshots:

  * gcc-4.0 may clone control flows, resulting in more sites

    with backpatchable calls. finalise_backpatching() now sets

    things up to exercise all affected control flow paths.

  * gcc-4.0 appears to ignore 'noinline' on static functions

    that are only called from one place, at least on x86-64.

    This broke perfctr_cpu_{write_control,isuspend,iresume}().

    Things work again if they are made non-static.

- Only define our own version of get_sb_pseudo() in kernels older

  than 2.6.11, since 2.6.11-rc1 added EXPORT_SYMBOL(get_sb_pseudo).

- In 2.6.11-rc2 and newer kernels, bind ioctls to ->unlocked_ioctl

  and ->compat_ioctl, and don't use register_ioctl32_conversion().

- Remove unused inode parameter to gperfctr_ioctl().

- Define static spinlocks with DEFINE_SPINLOCK(), following new

  coding style in 2.6.11-rc1. Add compat macros for older kernels.

Version 2.6.12, 2004-12-19

- PPC32 driver updated to be more robust in its detection of

  timebase and core clock frequencies. Some information sources

  can give wrong values for those frequencies, so the driver

  now tries other more reliable methods first.

Version 2.6.11, 2004-11-14

- Compat stuff updated for tsk->sighand->siglock,

  recalc_sigpending(), and preempt_enable_no_resched().

- Silence compiler warning from compat.h:remap_pfn_range().

- PPC32 overflow interrupt support backported from perfctr-2.7.

- Backported inheritance handling calls from perfctr-2.7

  to kernel patch kit. They are currently stubs, but can be

  implemented later without having to update the patch kit.

- Overflow interrupts fixes backported from perfctr-2.7.7:

  * x86/x86-64: move perfctr_suspend_thread() call from

    __switch_to() to the start of switch_to()

  * x86/x86-64: mask interrupts at suspend and record if

    any overflows are pending; unmask interrupts at resume

  * virtual: handle pending overflows in resume path

  * ppc32: provide dummy pending overflow checking function

Version 2.6.10.3, 2004-10-24

- virtual.c, linux/perfctr.h: reformatted "if( x )" to "if (x)"

  and similarly for while and switch statements.

- PPC32: Add support for MPC7447A. Add support for MPC7448,

  except for decoding its PLL_CFG.

- Move x86 cpu_type definitions from <linux/perfctr.h>

  to <asm-i386/perfctr.h>.

- Make PERFCTR_INTERRUPT_SUPPORT a Kconfig-derived option.

  Ditto PERFCTR_CPUS_FORBIDDEN_MASK_NEEDED.

  Also implement this in Config.in for 2.4 kernels.

- Kernel 2.6.10-rc1 removed the export of put_filp().

  Reordered the allocations in vperfctr_get_filp() to

  avoid the need to use put_filp().

- remap_page_range() was replaced with remap_pfn_range() in

  kernel 2.6.10-rc1. Updated virtual.c accordingly, and added

  remap_pfn_range() emulations for older kernels.

Version 2.6.10.2, 2004-10-19

- virtual.c: replace nrctrs_lock with a mutex. Avoids illegal

  may-sleep-while-holding-lock, caused by mutex operations in

  perfctr_cpu_{reserve,release}().

  Backport from perfctr-2.7.6.

- PPC32: Correct MMCR0 handling for FCECE/TRIGGER. Read

  MMCR0 at suspend and then freeze the counters. Move

  this code from read_counters() to suspend(). At resume,

  reload MMCR0 to unfreeze the counters. Clean up the

  cstatus checks controlling this behaviour.

  Backport from perfctr-2.7.6.

Version 2.6.10, 2004-09-14

- Fixed p4_clear_counters() to not access IQ_ESCR{0,1}

  on P4 models >= 3.

Version 2.6.10-pre1, 2004-08-03

- Changed x86-64 to use the x86 include file and driver.

  Intel's 64-bit P4 should now work in the x86-64 kernel.

- Replaced PERFCTR_INTERRUPT_SUPPORT and NMI_LOCAL_APIC

  #if:s in x86 code by #ifdef:s on CONFIG_X86_LOCAL_APIC.

- Use macros to clean up x86 per-cpu cache accesses.

- Recognize model 13 Pentium-Ms.

- Changed isuspend_cpu on x86 to be like x86-64's: it

  now stores a CPU number instead of a cache pointer.

- x86: make perfctr_cpu_name more approximate.

- The x86 driver records a simplified CPU type for x86_tests,

  but this only occurs if PERFCTR_INIT_TESTS is configured.

  perfctr_info.cpu_type is now unused.

- Changed P4 driver to set up and check an explicit flag

  for EXTENDED_CASCADE availability. perfctr_info.cpu_type

  is now unused except for perfctr_x86_init_tests().

- x86: Reformatted "if( x )" to "if (x)" and similarly for while

  and switch statements. Deleted #if 0 blocks.

Version 2.6.9, 2004-07-27

- Fix ppc_check_control() to allow 7400/7410 processors to

  specify MMCR2[THRESHMULT].

- PPC32 cleanups: make get_cpu_cache() return pointer not lvalue,

  eliminate duplicated initialisation/cleanup code.

- Makefile: enforce -fno-unit-at-a-time with gcc-3.4 on x86,

  to prevent stack overflow in 2.6 kernels < 2.6.6.

- Do sync_core() before rdtsc() in x86_tests, to avoid bogus

  benchmarking data on K8. Add sync_core() implementation for

  the 32-bit kernel. Add sync_core() benchmark.

- Added __perfctr_mk_cstatus() to allow x86.c:finalise_backpatching()

  to create a cstatus with i-mode counters marked as present, but

  with zero actual counters. This prevents perfctr_cpu_isuspend()

  from clearing the control register for counter #0 at init-time,

  when the hardware doesn't belong to this driver. On AMD and P6

  this would accidentally disable the NMI watchdog.

- x86: Marked initial targets of backpatchable calls

  'noinline' to prevent gcc from inlining them, which

  completely breaks the backpatching mechanism.

- x86_tests: fix CONFIG_X86_LOCAL_APIC=n linkage error.

- 2.6.8-rc1 no longer makes cpu_online_map a #define on UP,

  breaking modules. Reintroduce the macro.

- 2.6.8-rc1 changed cpus_complement() calling convention.

  Replace cpus_complement();cpus_and() with cpus_andnot(),

  and provide cpus_andnot() compat macro.

- PPC32: support generic CPUs using only the TB.

- PPC32: query OF for CPU/TB frequencies, drop /proc/cpuinfo

  parsing code.

- PPC32: avoid CPU re-detection in tests code.

- PPC32: clean up and sync with current perfctr-2.7 code.

Version 2.6.8, 2004-05-29

- Added recognition of PowerPC 750GX.

- Changes for the {reserve,release}_lapic_nmi() API added in

  kernel 2.6.6 backported from perfctr-2.7.1:

  * Starting with kernel 2.6.6 we no longer need access to

    nmi_perfctr_msr, so removed EXPORT_SYMBOL() and <asm/apic.h>

    patches related to this variable (except for older kernels).

  * Updated x86.c to use the new API. Added simulation (without

    the non-conflict guarantees) for older kernels.

  * Moved hardware reservation to x86.c's "reserve" procedure.

    The init code now only does read-only hardware detection.

  * Added a mutex to the reserve/release procedures, eliminating

  * a long-standing race possibility.

  * Changed x86.c to reserve and release the hardware around its

    call to perfctr_x86_init_tests().

  * Similarly updated x86_64.c for the new API.

Version 2.6.7, 2004-05-04

- Replaced x86_64_tests.{c,h} with x86_tests.{c,h}.

- sys_device_{,un}register() was renamed as sysdev_{,un}register()

  in 2.6.4-rc2. Updated x86.c and x86_64.c accordingly, and

  added a compatibility definition in compat.h.

- Removed unnecessary '#include "compat.h"' from x86_tests.c.

- Replaced x86_64_setup.c with x86_setup.c.

- Replaced x86_64_compat.h with x86_compat.h.

- Moved perfctr_interrupt entry point from x86_setup.c to patch kit,

  for kernels older than 2.4.21. Cleanup to facilitate future merge

  of x86_setup.c and x86_64_setup.c.

Version 2.6.6, 2004-02-21

- Fixed a bug in x86-64's perfctr interrupt entry code in 2.4 kernels,

  causing it to pass the wrong value for "struct pt_regs*". This

  was harmless since the retrieved "rip" was unused, but still wrong.

  Renamed do_perfctr_interrupt to smp_perfctr_interrupt to allow

  using the 2.4 kernel's standard BUILD_SMP_INTERRUPT macro.

- Unmask LVTPC after interrupt on Pentium-M. An oprofile user

  reports that P-M auto-masks LVTPC just like P4. Preliminary

  measurements indicate a 40 to 60 cycle cost for the apic write

  on P4s and P6s, so the unmask is not done unconditionally.

- Measure LVTPC write overhead in x86{,_64}_tests.c.

- Add Pentium 4 Model 3 detection.

- The 2.4.21-193 SuSE kernel does EXPORT_SYMBOL(mmu_cr4_features).

  Add compat24.h workaround for this.

Version 2.6.5, 2004-01-26

- Added perfctr_info.cpu_type constants to <asm-ppc/perfctr.h>.

- Init filp->f_mapping in virtual.c for 2.6.2-rc1+ kernels.

- Updated p4_check_control():

  * Allow ESCR.CPL_T1 to be non-zero when using global-mode

    counters on HT processors.

  * Don't require ESCR.CPL_T0 to be non-zero. CPL_T0==0b00

    is safe and potentially useful (global counters on HT).

  * Require CCCR.ACTIVE_THREAD==0b11 on non-HT processors, as

    documented in the IA32 Volume 3 manual. Old non-HT P4s

    seem to work Ok for all four values (see perfctr-2.6.0-pre3

    notes), but this is neither guaranteed nor useful.

- x86.c now detects & records P4 HT-ness also in UP kernels.

- Added 'is_global' parameter to perfctr_cpu_update_control().

  This flag is ignored on everything except P4 (sigh).

Version 2.6.4, 2004-01-12

- Added 'tsc_to_cpu_mult' field to struct perfctr_info, replacing

  '_reserved1'. This is needed on PowerPC to map time-base ticks

  to actual time. On x86/AMD64, tsc_to_cpu_mult == 1.

- Added support for PowerPC 604/7xx/74xx processors. Overflow

  interrupts are currently not allowed due to the PMI/DECR erratum.

- Replaced perfctr_cpus_mask() with cpus_addr(). Updated cpumask.h

  to define cpus_addr() for kernels older than 2.6.1.

Version 2.6.3-pl1, 2004-01-01

- Moved the x86 interrupt handler definition from x86_setup.c to

  the patch kit for 2.4.21 and later 2.4 kernels, like it already

  is done for 2.6 kernels. This change is needed due to extensive

  interrupt handler changes in RedHat's 2.4.21-6.EL kernel.

- Simplified <asm-i386/perfctr.h>: now that early 2.4 kernels no

  longer are supported, LOCAL_PERFCTR_VECTOR is known to be defined,

  so CONFIG_X86_LOCAL_APIC implies PERFCTR_INTERRUPT_SUPPORT.

Version 2.6.3, 2003-12-21

- Removed gperfctr_cpu_state_only_cpu_sdesc's total_sizeof

  optimisation. The ABI change in 2.6.2 broke it, leading to

  the new fields not being cleared and later causing EOVERFLOW.

- The perfctr_ioctl32_handler() workaround is now only applied

  to kernels older than 2.4.23, since 2.4.23 added the "NULL

  handler == sys_ioctl" logic.

Version 2.6.2, 2003-11-23

- Added 16 bytes (four fields) of reserved data to perfctr_info,

  perfctr_cpu_control, vperfctr_control, gperfctr_cpu_control,

  and gperfctr_cpu_state. Renumbered marshalling tags for

  generic structures. Bumped ABI versions.

- Only allow use of IQ_ESCR{0,1} on P4 models <= 2. These ESCRs

  were removed from later models, according to a recent Intel

  documentation update (252046-006).

- Fixes for Fedora Core 1's 2.4.22-1.2115.nptl kernel:

  * Work around their incomplete and broken cpumask_t backport.

  * Avoid name conflict due to their on_each_cpu() backport.

  * Handle their preempt_disable()/enable() macros.

- Added new perfctr_cpu_is_forbidden() macro to fix a

  compilation error affecting AMD64 in SMP 2.6 kernels.

  SMP cpu_isset() requires that mask is an lvalue, but

  for AMD64 the mask is a constant.

Version 2.6.1, 2003-10-05

- Kernel 2.6.0-test6 changed /proc/self and the /proc/<pid>/

  namespace to refer to "processes" (groups of CLONE tasks)

  instead of actual kernel tasks. This forced the planned

  transition of the vperfctr API from /proc/<pid>/perfctr

  to /dev/perfctr to occur immediately. Changes:

  * Moved /dev/perfctr implementation from global.c to init.c.

  * Implemented VPERFCTR_{CREAT,OPEN}, vperfctr_attach(), and

    the vperfctrfs pseudo-fs needed to support the magic files.

    The fs code was ported from perfctr-1.6/3.1, but updated

    for 2.6 and fixed to permit module unloading in 2.4.

  * Fixed VPERFCTR_OPEN to accept tsk->thread.perfctr == NULL.

    (Needed to info querying commands.)

  * Removed /proc/<pid>/perfctr code. Simplified vperfctr_stub code.

  * Updated vperfctr_attach() to mimic the old /proc vperfctr_open().

    This fixes some synchronisation issues.

- Cleanups:

  * Removed #if checks and code for kernels older than 2.4.16.

  * Eliminated compat macros that are identical in 2.6 and 2.4.

  * Moved ptrace_check_attach EXPORT_SYMBOL from x86{,_64}_setup.c

    to virtual_stub.c.

  * get_task_by_proc_pid_inode() is now trivial. Eliminated it.

  * p4_ht_finalise() is now trivial. Eliminated it.

- Added MODULE_ALIAS() declaration, eliminating the need for

  an alias in /etc/modprobe.conf with 2.6 kernels. Added

  MODULE_ALIAS() compatibility #define in compat24.h.

- Added detection of AMD K8 Revision C processors.

- Updated K8C detection for Revision C Athlon64s.

Version 2.6.0, 2003-09-08

- Handle set_cpus_allowed() when PERFCTR_CPUS_FORBIDDEN_MASK_NEEDED:

  * Add bad_cpus_allowed flag to struct vperfctr.

  * Check bad_cpus_allowed in __vperfctr_resume: if resuming

    with PMCs on forbidden CPU, kill counters and SIGILL current.

  * __vperfctr_set_cpus_allowed() callback: set bad_cpus_allowed

    and print warning if mask allows forbidden CPUs.

  * Use task_lock/unlock instead of preempt_disable/enable to

    synchronise task_struct accesses.

  * Ensure sampling_timer and bad_cpus_allowed share cache line.

  * #include <linux/compiler.h> explicitly for 2.4.18 and older

    kernels; newer kernels include it from <linux/kernel.h>.

  * Hook in virtual_stub.c.

  * Hook and cpumask_t typedef in <linux/perfctr.h>.

- Simplify #if test for set_cpus_allowed() emulation code.

  Also don't define it if CONFIG_PERFCTR_VIRTUAL isn't set.

- cpumask.h only typedefs cpumask_t if <linux/perfctr.h> hasn't.

- Don't hide #include <linux/kernel.h> in compat24.h.

- Fixed compat24.h to test for MODULE not CONFIG_MODULES at the

  __module_get/module_put macros.

Version 2.6.0-pre5, 2003-08-31

- printk() is not allowed in switch_to(). Disabled debug code

  which could violate that rule. Changed virtual_stub.c to BUG()

  instead of printk() if the driver is invoked when not loaded.

- Renamed vperfctr_exit2() to vperfctr_unlink() for clarity.

- gcc-3.3.1 issued several "dereferencing type-punned pointer will

  break strict-aliasing rules" warnings for marshal.c. Used explicit

  unions to fix the warnings and clean up the code.

- Removed compat22.h.

- cpumask_t was included in standard 2.6.0-test4; replace #ifndef

  test in cpumask.h with normal kernel version test.

- x86-64 fix: sys_ioctl() isn't exported to modules, so call

  filp->f_op->ioctl() instead in perfctr_ioctl32_handler().

- x86-64 fix: init.c must include <asm/ioctl32.h> not <linux/ioctl32.h>

  for compatibility with 2.4 kernels.

Version 2.6.0-pre4, 2003-08-19

- Fix x86-64 register_ioctl32_conversion() usage for 2.4 kernels:

  * Supply dummy handler since a NULL handler oopses the kernel.

  * Test CONFIG_IA32_EMULATION since CONFIG_COMPAT is post-2.4.

- Fixed and merged the new API struct marshalling code:

  * New files marshal.c and marshal.h contain the marshalling code

    and high-level helper functions (source shared with the library).

  * User-space structs are struct perfctr_struct_buf and accessed using

    perfctr_copy_{from,to}_user() with ptr to appropriate descriptor.

    The cpumask stuff isn't changed.

  * All ioctls registered as trivially 32-bit compatible on x86-64.

  * Changed perfctr_info cpu_type/cpu_features from short to int:

    this avoids the need for UINT16 marshalling support, and cpumask_t

    caused perfctr_info to change binary representation anyway.

- Declared VPERFCTR_{CREAT,OPEN} ioctls, but left them unimplemented.

- Fixed vperfctr_open() preemption bug. The O_CREAT check+install

  code could be preempted, leading to remote-control races.

- Fixed perfctr_exit_thread() preemption bug. It detached the vperfctr

  before calling __vperfctr_exit(). If current was preempted before

  __vperfctr_exit() called vperfctr_suspend(), perfctr_suspend_thread()

  would fail to suspend the counters. The suspend+detach is now done

  atomically within __vperfctr_exit().

- Changes to handle 2.6 kernels with the cpumask_t patch (-mm, -osdl):

  * Convert perfctr_cpus_forbidden_mask accesses to cpumask_t API.

    Based in part on a patch for the -osdl kernel by Stephen Hemminger.

  * Remove cpus and cpus_forbidden from struct perfctr_info,

    since their sizes depend on the kernel configuration.

  * Add struct perfctr_cpu_mask to export cpumask_t objects

    sanely (i.e., using ints not longs) to user-space.

  * Add CPUS and CPUS_FORBIDDEN commands to retrieve these sets.

  * Add cpumask.h to emulate cpumask_t API in cpumask_t-free kernels.

  * Move perfctr_cpus_forbidden_mask declaration/#define from

    <asm/perfctr.h> to cpumask.h -- necessary since <asm/perfctr.h>

    doesn't have access to the driver's compatibility definitions.

- Cleaned up perfctr_cpu_ireload().

- Removed struct field offset check from init.c.

- 2.4.22-rc1 does EXPORT_SYMBOL(mmu_cr4_features). Added

  new compat #define to handle this.

- Rename x86.c's rdmsrl() to rdmsr_low() to work around msr.h

  changes in 2.6.0-test3. Also rename rdpmcl() to rdpmc_low().

- Replaced __attribute__((__aligned__(SMP_CACHE_BYTES))) usage

  with the official ____cacheline_aligned macro.

- Detect cpuid 0x69x VIA C3s (Antaur/Nehemiah).

Version 2.6.0-pre3, 2003-08-03

- Changed perfctr_info.cpus and cpus_forbidden to be int instead of

  long, to make x86-32 and x86-64 compatible. This is a temporary

  solution, as there are patches for >32 CPUs on x86-32. The real

  solution is to make these sets variable-sized, and have user-space

  retrieve them with a new command.

- Simplified GPERFCTR_CONTROL to update a single CPU instead of

  a set of CPUs. Moved cstatus clearing to release_hardware().

- Moved gperfctr start to new GPERFCTR_START command.

- Simplified GPERFCTR_READ to access a single CPU instead of a

  set of CPUs.

- Removed the requirement that CCCR.ACTIVE_THREAD == 3 on P4.

  HT processors define behaviour for all four possible values,

  and non-HT processors behave sanely for all four values.

- Moved struct perfctr_low_ctrs definition from <asm/perfctr.h> to

  the corresponding low-level driver, since it's only used there.

- Changed perfctr_info.cpu_khz and vperfctr_control.preserve to be

  int instead of long. This corrects x86-64 and makes it compatible

  with x86-32.

- Updated x86.c to permit extended cascading on P4M2.

- Fixed a bug where the perfctr module's refcount could be zero with

  code still running in the module (pending returns to exit_thread()).

  This could race with rmmod in preemptive kernels, and in theory

  also in SMP kernels.

  * module owner field added to vperfctr_stub

  * _vperfctr_exit() in the modular case is now a function in

    vperfctr_stub.c, which brackets the vperfctr_stub.exit() call

    with __module_get() and module_put() on vperfctr_stub.owner

  * updated 2.4 and 2.2 compat definitions of __module_get() and

    module_put() to work for modules != THIS_MODULE  

- Replaced uses of (void)try_module_get() with __module_get() as the

  latter is more appropriate for 2.6 kernels. Updated compat stuff.

Version 2.6.0-pre2, 2003-07-13

- vperfctr API fixes:

  * The new VPERFCTR_READ_CONTROL command retrieves a vperfctr's

    control data.

  * Renamed VPERFCTR_SAMPLE to VPERFCTR_READ_SUM, and made it

    write the sums to a perfctr_sum_ctrs user-space buffer.

  * Non-write commands are now always permitted on unlinked perfctrs.

  The first change was needed since the control data no longer is

  accessible via the mmap()ed state. The other changes clean up and

  simplify perfex and the library's slow-path read_ctrs() operation.

- sys_vperfctr_ functions now mark the tsk parameter as "const" if

  they don't need write access to it. Typically they only need to

  compare it with current to detect self-access cases.

- perfctr_cpu_state no longer makes the perfctr_cpu_control part

  accessible to user-space (via mmap() of vperfctrs).

- Simplified {set,is}_isuspend_cpu() in x86_64.c by having callers

  pass the CPU number instead of the cache pointer (which was only

  used to derive the CPU number).

- Eliminated NMI_LOCAL_APIC #ifs from x86-64 code since x86-64

  always defines it.

- x86.c cleanups: the non-PERFCTR_INTERRUPT_SUPPORT case now uses

  dummy stub functions, eliminated six #ifdefs.

- x86_64_setup.c needs <asm/fixmap.h>.

- Protected cpu_has_mmx and cpu_has_ht #defines in x86_compat.h

  with #ifndef since 2.4.22-pre3 added those #defines.

- Eliminated PERFCTR_INTERRUPT_SUPPORT #ifs from x86-64 code

  since x86-64 always defines CONFIG_X86_LOCAL_APIC.

- Removed the P4-specific versions of isuspend() and iresume().

  P4 now uses p6_like_{isuspend,iresume}(), just like P6/K7/K8.

- Long overdue cleanup in x86.c/x86_64.c: renamed per_cpu_cache

  pointer variables from 'cpu' to 'cache'.

- Added inline functions in virtual.c for registering the overflow

  handler and for clearing iresume_cstatus. Cleaned out several

  #if PERFCTR_INTERRUPT_SUPPORT occurrences from the main code.

  (Partial backport from the abandoned perfctr-3.1 branch.)

- Inlined now useless 'struct vperfctr_state' in 'struct vperfctr'.

Version 2.6.0-pre1, 2003-07-02

- Rearranged 'struct perfctr_cpu_state' to reduce the number of

  cache lines needed to be touched by key operations (suspend,

  resume, sample). Switched from struct-of-arrays to array-of-struct

  for perfctr counts, and copied pmc_map into the PMC data array.

  The old representation touched at least 3 cache lines at key

  operations, the new one only needs one cache line in most cases.

  The user-space mmap() view of the new representation is binary

  compatible between x86 and x86-64.

- Changed 'isuspend_cpu' in perfctr_cpu_state on x86-64 to be a

  32-bit CPU number, to maintain binary compatibility with x86.

- Removed the union of p5_cesr and id; use id throughout.

- Removed _filler and si_signo from 'struct vperfctr_state', making

  the user-space view of it identical to 'struct perfctr_cpu_state'.

Version 2.5.5, 2003-06-15

- Updated x86 driver for 2.5.71 local APIC driver model changes.

- Updated x86-64 driver for 2.5.71 NMI watchdog enable/disable API.

- x86-64 is broken in 2.5.71 since x86-64 updated to driver model

  for local APIC and NMI watchdog, at the same time as x86 moved

  to a newer version of the "system device" driver model. Updated

  the x86-64 driver for the new model, which is expected to be in

  x86-64 by 2.5.72 (patch exists for 2.5.71).

Version 2.5.4, 2003-06-01

- The generic-x86-with-TSC driver now uses rdpmc_read_counters

  and p6_write_control instead of its own procedures.

- K8 docs are now available. Updated comment in x86.c accordingly.

- P4 OVF_PMI+FORCE_OVF counters didn't work at all, resulting in

  BUG messages from the driver since identify_overflow failed to

  detect which counters had overflowed, and vperfctr_ihandler

  left the vperfctr in an inconsistent state. This works now.

  However, hardware quirks makes this configuration only useful

  for one-shot counters, since resuming generates a new interrupt

  and the faulting instruction again doesn't complete. The same

  problem can occur with regular OVF_PMI counters if ireset is

  a small-magnitude value, like -5.

  This is a user-space problem; the driver survives.

- On P4, OVF_PMI+FORCE_OVF counters must have an ireset value of -1.

  This allows the regular overflow check to also handle FORCE_OVF

  counters. Not having this restriction would lead to MAJOR

  complications in the driver's "detect overflow counters" code.

  There is no loss of functionality since the ireset value doesn't

  affect the counter's PMI rate for FORCE_OVF counters.

- Moved P4 APIC_LVTPC reinit from p4_isuspend() to identify_overflow().

  Reduces context-switch overheads when i-mode counters are active.

- Corrected vperfctr_suspend()'s precondition.

- Corrected comment in <asm/perfctr.h> to state that ireset[]

  values must be negative rather than non-positive.

- Made 'perfctr_cpu_name' __initdata, like its predecessor.

Version 2.5.3.1, 2003-05-21

- Replaced 'char *perfctr_cpu_name[]' by 'char *perfctr_cpu_name'.

  This is needed for x86-64 and other non-x86 architectures.

- Changed <asm-x86_64/perfctr.h> to use 'long long' for 64-bit sums.

  This doesn't change the ABI, but improves user-space source code

  compatibility with 32-bit x86.

- Removed the !defined(set_cpus_allowed) check added to compat24.h

  in 2.5.3. It's wrong for SMP builds with modules and MODVERSIONS,

  since the set_cpus_allowed() emulation function becomes a #define

  from include/linux/modules/x86_setup.ver. Instead add the already

  used HAVE_SET_CPUS_ALLOWED #define to include/linux/config.h in

  the kernel patch, but make it conditional on CONFIG_X86_64.

Version 2.5.3, 2003-05-16

- Added detection code for Pentium M. MISC_ENABLE_PERF_AVAIL is

  now checked on both P4 and Pentium M.

- Added x86_64 driver code. Both x86_64.c and asm-x86_64/perfctr.h

  are basically simplified versions of corresponding x86 files,

  with P5 and P4 support removed, 2.2 kernel support removed, and

  'long long' for sums replaced by 'long'. The last change is

  painful for user-space and may be reverted.

- compat24.h: don't define set_cpus_allowed() if already #defined,

  workaround for RawHide's 2.4.20-9.2 x86_64 kernel.

- Removed list of supported CPUs from Kconfig. That information

  belongs elsewhere (and it's a pain to maintain for 2.2/2.4).

Version 2.5.2, 2003-04-13

- Minor cleanup: use PROC_I() unconditionally in virtual.c,

  implement trivial compat macro in compat24.h.

- Updated power management code for the local APIC and NMI

  watchdog driver model changes in kernel 2.5.67.

  The suspend/resume procedures are still no-ops, however.

  This revealed a bug in the lapic_nmi_watchdog resume code:

  it resumes the lapic_nmi_watchdog even when it was disabled

  before suspend. Perfctr's 2.5.67 kernel patch includes a fix.

- perfctr_sample_thread() is now used also on UP. Anton Ertl's

  2.26GHz UP P4 managed to execute a process for more than 2^32

  cycles before suspending it, causing TSC inaccuracies.

- RH9's 2.4.20-8 kernel changed cpu_online(), put_task_struct() and

  remap_page_range() to be more like in 2.5 kernels, and moved the

  declaration of ptrace_check_attach() from mm.h to ptrace.h, also

  like in 2.5 kernels, requiring fixes to compat24.h and x86_setup.c.

- Added note in x86.c about the new Pentium M processor.

Version 2.5.1, 2003-03-23

- Fix P4 HT initialisation. I've seen several boot logs from

  people running MP P4 Xeons with HT disabled: this produces

  an ugly "restricting access for CPUs 0x0" message, and would

  cause P4 HT init to unnecessarily return error in older kernels

  lacking set_cpus_allowed(). Now only print the message or

  signal error if non-zero siblings actually are found.

- The set_cpus_allowed() emulation doesn't compile in 2.4

  kernels older than 2.4.15 due to the p->cpus_running field.

  Updated version checks to skip it in 2.4.x when x<15.

- Fix set_cpus_allowed() emulation compile error on BUG_ON()

  in 2.4 kernels older than 2.4.19.

- Added Nehemiah note/reminder in x86.c:centaur_init().

Version 2.5.0, 2003-03-10

- Reverted the 2.5.0-pre2 change that replaced the PERFCTR_INFO

  ioctl by read(): it made the API look too weird.

  Added a PERFCTR_ABI ioctl which only retrieves 'abi_version'.

- Cleaned up struct perfctr_info: renamed abi_magic to abi_version,

  and version to driver_version. Renamed PERFCTR_*_MAGIC too.

- Cleaned up struct perfctr_cpu_control: moved evntsel_aux[]

  into the p4 sub-struct and renamed it as escr[]. Only P4 needs

  it anyway, and the new name clarifies its purpose.

- Renumbered the vperfctr ioctls to the 8-15 range (8-11 are used)

  and reserved 0-7 (0-1 are used) for generic ioctls.

- Added 'use_nmi' field to struct gperfctr_control, reserved for

  future use if/when support for i-mode gperfctrs is implemented.

- Replaced some preempt/smp_call_function combinations with 2.5.64's

  new on_each_cpu() construct. Added compatibility definitions to

  compat24.h and compat22.h.

Version 2.5.0-pre2, 2003-03-03

- Added ABI version to perfctr_info. Replaced PERFCTR_INFO ioctl

  by read() on the fd, since that allows reading the ABI version

  even in the case of a version mismatch. Removed binary layout

  magic number from vperfctr_state. Rearranged perfctr_info to

  make the 'long' fields 8-byte aligned.

- Added #ifdef CONFIG_KPERFCTR to <linux/perfctr.h> to ensure

  that <asm/perfctr.h> isn't included unless CONFIG_KPERFCTR=y.

  This allows the patched kernel source to compile cleanly also

  in archs not yet supported by perfctr.

- Removed PERFCTR_PROC_PID_MODE #define and replaced it with

  /*notype*/S_IRUSR in the patch files.

- Added perfctr_vector_init() to <asm-i386/perfctr.h>. Cleaned

  up arch/i386/kernel/i8259.c patch.

- Removed apic_lvtpc_irqs[] array. Removed irq.c patch.

- Updated CONFIG_PERFCTR_INIT_TESTS help text to match reality.

- Kernel 2.4.21-pre5 added set_cpus_allowed(), which required

  fixing compat24.h and x86_setup.c.

- Fixed init.c for kernel 2.5.63 removing EXPORT_NO_SYMBOLS.

- Cleaned up compat.h by moving 2.2/2.4 stuff to separate files.

Version 2.5.0-pre1, 2003-02-19

- Repair global perfctr API: the target CPUs are now explicit

  in the calls to write control and read state. Global perfctrs

  now work on 2.5 SMP kernels (which no longer have smp_num_cpus