In preparation to enable user counter access on arm64 and to move some
of the user access handling to perf core, create a common event flag for
user counter access and convert x86 to use it.
Since the architecture specific flags start at the LSB, starting at the
MSB for common flags.
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Borislav Petkov <bp@alien8.de>
Cc: x86@kernel.org
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: linux-perf-users@vger.kernel.org
Reviewed-by: Mark Rutland <mark.rutland@arm.com>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/r/20211208201124.310740-2-robh@kernel.org
Signed-off-by: Will Deacon <will@kernel.org>
- Remove socket skb caches
- Add a SO_RESERVE_MEM socket op to forward allocate buffer space
and avoid memory accounting overhead on each message sent
- Introduce managed neighbor entries - added by control plane and
resolved by the kernel for use in acceleration paths (BPF / XDP
right now, HW offload users will benefit as well)
- Make neighbor eviction on link down controllable by userspace
to work around WiFi networks with bad roaming implementations
- vrf: Rework interaction with netfilter/conntrack
- fq_codel: implement L4S style ce_threshold_ect1 marking
- sch: Eliminate unnecessary RCU waits in mini_qdisc_pair_swap()
BPF:
- Add support for new btf kind BTF_KIND_TAG, arbitrary type tagging
as implemented in LLVM14
- Introduce bpf_get_branch_snapshot() to capture Last Branch Records
- Implement variadic trace_printk helper
- Add a new Bloomfilter map type
- Track <8-byte scalar spill and refill
- Access hw timestamp through BPF's __sk_buff
- Disallow unprivileged BPF by default
- Document BPF licensing
Netfilter:
- Introduce egress hook for looking at raw outgoing packets
- Allow matching on and modifying inner headers / payload data
- Add NFT_META_IFTYPE to match on the interface type either from
ingress or egress
Protocols:
- Multi-Path TCP:
- increase default max additional subflows to 2
- rework forward memory allocation
- add getsockopts: MPTCP_INFO, MPTCP_TCPINFO, MPTCP_SUBFLOW_ADDRS
- MCTP flow support allowing lower layer drivers to configure msg
muxing as needed
- Automatic Multicast Tunneling (AMT) driver based on RFC7450
- HSR support the redbox supervision frames (IEC-62439-3:2018)
- Support for the ip6ip6 encapsulation of IOAM
- Netlink interface for CAN-FD's Transmitter Delay Compensation
- Support SMC-Rv2 eliminating the current same-subnet restriction,
by exploiting the UDP encapsulation feature of RoCE adapters
- TLS: add SM4 GCM/CCM crypto support
- Bluetooth: initial support for link quality and audio/codec
offload
Driver APIs:
- Add a batched interface for RX buffer allocation in AF_XDP
buffer pool
- ethtool: Add ability to control transceiver modules' power mode
- phy: Introduce supported interfaces bitmap to express MAC
capabilities and simplify PHY code
- Drop rtnl_lock from DSA .port_fdb_{add,del} callbacks
New drivers:
- WiFi driver for Realtek 8852AE 802.11ax devices (rtw89)
- Ethernet driver for ASIX AX88796C SPI device (x88796c)
Drivers:
- Broadcom PHYs
- support 72165, 7712 16nm PHYs
- support IDDQ-SR for additional power savings
- PHY support for QCA8081, QCA9561 PHYs
- NXP DPAA2: support for IRQ coalescing
- NXP Ethernet (enetc): support for software TCP segmentation
- Renesas Ethernet (ravb) - support DMAC and EMAC blocks of
Gigabit-capable IP found on RZ/G2L SoC
- Intel 100G Ethernet
- support for eswitch offload of TC/OvS flow API, including
offload of GRE, VxLAN, Geneve tunneling
- support application device queues - ability to assign Rx and Tx
queues to application threads
- PTP and PPS (pulse-per-second) extensions
- Broadcom Ethernet (bnxt)
- devlink health reporting and device reload extensions
- Mellanox Ethernet (mlx5)
- offload macvlan interfaces
- support HW offload of TC rules involving OVS internal ports
- support HW-GRO and header/data split
- support application device queues
- Marvell OcteonTx2:
- add XDP support for PF
- add PTP support for VF
- Qualcomm Ethernet switch (qca8k): support for QCA8328
- Realtek Ethernet DSA switch (rtl8366rb)
- support bridge offload
- support STP, fast aging, disabling address learning
- support for Realtek RTL8365MB-VC, a 4+1 port 10M/100M/1GE switch
- Mellanox Ethernet/IB switch (mlxsw)
- multi-level qdisc hierarchy offload (e.g. RED, prio and shaping)
- offload root TBF qdisc as port shaper
- support multiple routing interface MAC address prefixes
- support for IP-in-IP with IPv6 underlay
- MediaTek WiFi (mt76)
- mt7921 - ASPM, 6GHz, SDIO and testmode support
- mt7915 - LED and TWT support
- Qualcomm WiFi (ath11k)
- include channel rx and tx time in survey dump statistics
- support for 80P80 and 160 MHz bandwidths
- support channel 2 in 6 GHz band
- spectral scan support for QCN9074
- support for rx decapsulation offload (data frames in 802.3
format)
- Qualcomm phone SoC WiFi (wcn36xx)
- enable Idle Mode Power Save (IMPS) to reduce power consumption
during idle
- Bluetooth driver support for MediaTek MT7922 and MT7921
- Enable support for AOSP Bluetooth extension in Qualcomm WCN399x
and Realtek 8822C/8852A
- Microsoft vNIC driver (mana)
- support hibernation and kexec
- Google vNIC driver (gve)
- support for jumbo frames
- implement Rx page reuse
Refactor:
- Make all writes to netdev->dev_addr go thru helpers, so that we
can add this address to the address rbtree and handle the updates
- Various TCP cleanups and optimizations including improvements
to CPU cache use
- Simplify the gnet_stats, Qdisc stats' handling and remove
qdisc->running sequence counter
- Driver changes and API updates to address devlink locking
deficiencies
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
-----BEGIN PGP SIGNATURE-----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=srde
-----END PGP SIGNATURE-----
Merge tag 'net-next-for-5.16' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next
Pull networking updates from Jakub Kicinski:
"Core:
- Remove socket skb caches
- Add a SO_RESERVE_MEM socket op to forward allocate buffer space and
avoid memory accounting overhead on each message sent
- Introduce managed neighbor entries - added by control plane and
resolved by the kernel for use in acceleration paths (BPF / XDP
right now, HW offload users will benefit as well)
- Make neighbor eviction on link down controllable by userspace to
work around WiFi networks with bad roaming implementations
- vrf: Rework interaction with netfilter/conntrack
- fq_codel: implement L4S style ce_threshold_ect1 marking
- sch: Eliminate unnecessary RCU waits in mini_qdisc_pair_swap()
BPF:
- Add support for new btf kind BTF_KIND_TAG, arbitrary type tagging
as implemented in LLVM14
- Introduce bpf_get_branch_snapshot() to capture Last Branch Records
- Implement variadic trace_printk helper
- Add a new Bloomfilter map type
- Track <8-byte scalar spill and refill
- Access hw timestamp through BPF's __sk_buff
- Disallow unprivileged BPF by default
- Document BPF licensing
Netfilter:
- Introduce egress hook for looking at raw outgoing packets
- Allow matching on and modifying inner headers / payload data
- Add NFT_META_IFTYPE to match on the interface type either from
ingress or egress
Protocols:
- Multi-Path TCP:
- increase default max additional subflows to 2
- rework forward memory allocation
- add getsockopts: MPTCP_INFO, MPTCP_TCPINFO, MPTCP_SUBFLOW_ADDRS
- MCTP flow support allowing lower layer drivers to configure msg
muxing as needed
- Automatic Multicast Tunneling (AMT) driver based on RFC7450
- HSR support the redbox supervision frames (IEC-62439-3:2018)
- Support for the ip6ip6 encapsulation of IOAM
- Netlink interface for CAN-FD's Transmitter Delay Compensation
- Support SMC-Rv2 eliminating the current same-subnet restriction, by
exploiting the UDP encapsulation feature of RoCE adapters
- TLS: add SM4 GCM/CCM crypto support
- Bluetooth: initial support for link quality and audio/codec offload
Driver APIs:
- Add a batched interface for RX buffer allocation in AF_XDP buffer
pool
- ethtool: Add ability to control transceiver modules' power mode
- phy: Introduce supported interfaces bitmap to express MAC
capabilities and simplify PHY code
- Drop rtnl_lock from DSA .port_fdb_{add,del} callbacks
New drivers:
- WiFi driver for Realtek 8852AE 802.11ax devices (rtw89)
- Ethernet driver for ASIX AX88796C SPI device (x88796c)
Drivers:
- Broadcom PHYs
- support 72165, 7712 16nm PHYs
- support IDDQ-SR for additional power savings
- PHY support for QCA8081, QCA9561 PHYs
- NXP DPAA2: support for IRQ coalescing
- NXP Ethernet (enetc): support for software TCP segmentation
- Renesas Ethernet (ravb) - support DMAC and EMAC blocks of
Gigabit-capable IP found on RZ/G2L SoC
- Intel 100G Ethernet
- support for eswitch offload of TC/OvS flow API, including
offload of GRE, VxLAN, Geneve tunneling
- support application device queues - ability to assign Rx and Tx
queues to application threads
- PTP and PPS (pulse-per-second) extensions
- Broadcom Ethernet (bnxt)
- devlink health reporting and device reload extensions
- Mellanox Ethernet (mlx5)
- offload macvlan interfaces
- support HW offload of TC rules involving OVS internal ports
- support HW-GRO and header/data split
- support application device queues
- Marvell OcteonTx2:
- add XDP support for PF
- add PTP support for VF
- Qualcomm Ethernet switch (qca8k): support for QCA8328
- Realtek Ethernet DSA switch (rtl8366rb)
- support bridge offload
- support STP, fast aging, disabling address learning
- support for Realtek RTL8365MB-VC, a 4+1 port 10M/100M/1GE switch
- Mellanox Ethernet/IB switch (mlxsw)
- multi-level qdisc hierarchy offload (e.g. RED, prio and shaping)
- offload root TBF qdisc as port shaper
- support multiple routing interface MAC address prefixes
- support for IP-in-IP with IPv6 underlay
- MediaTek WiFi (mt76)
- mt7921 - ASPM, 6GHz, SDIO and testmode support
- mt7915 - LED and TWT support
- Qualcomm WiFi (ath11k)
- include channel rx and tx time in survey dump statistics
- support for 80P80 and 160 MHz bandwidths
- support channel 2 in 6 GHz band
- spectral scan support for QCN9074
- support for rx decapsulation offload (data frames in 802.3
format)
- Qualcomm phone SoC WiFi (wcn36xx)
- enable Idle Mode Power Save (IMPS) to reduce power consumption
during idle
- Bluetooth driver support for MediaTek MT7922 and MT7921
- Enable support for AOSP Bluetooth extension in Qualcomm WCN399x and
Realtek 8822C/8852A
- Microsoft vNIC driver (mana)
- support hibernation and kexec
- Google vNIC driver (gve)
- support for jumbo frames
- implement Rx page reuse
Refactor:
- Make all writes to netdev->dev_addr go thru helpers, so that we can
add this address to the address rbtree and handle the updates
- Various TCP cleanups and optimizations including improvements to
CPU cache use
- Simplify the gnet_stats, Qdisc stats' handling and remove
qdisc->running sequence counter
- Driver changes and API updates to address devlink locking
deficiencies"
* tag 'net-next-for-5.16' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next: (2122 commits)
Revert "net: avoid double accounting for pure zerocopy skbs"
selftests: net: add arp_ndisc_evict_nocarrier
net: ndisc: introduce ndisc_evict_nocarrier sysctl parameter
net: arp: introduce arp_evict_nocarrier sysctl parameter
libbpf: Deprecate AF_XDP support
kbuild: Unify options for BTF generation for vmlinux and modules
selftests/bpf: Add a testcase for 64-bit bounds propagation issue.
bpf: Fix propagation of signed bounds from 64-bit min/max into 32-bit.
bpf: Fix propagation of bounds from 64-bit min/max into 32-bit and var_off.
net: vmxnet3: remove multiple false checks in vmxnet3_ethtool.c
net: avoid double accounting for pure zerocopy skbs
tcp: rename sk_wmem_free_skb
netdevsim: fix uninit value in nsim_drv_configure_vfs()
selftests/bpf: Fix also no-alu32 strobemeta selftest
bpf: Add missing map_delete_elem method to bloom filter map
selftests/bpf: Add bloom map success test for userspace calls
bpf: Add alignment padding for "map_extra" + consolidate holes
bpf: Bloom filter map naming fixups
selftests/bpf: Add test cases for struct_ops prog
bpf: Add dummy BPF STRUCT_OPS for test purpose
...
- Cleanup of extable fixup handling to be more robust, which in turn
allows to make the FPU exception fixups more robust as well.
- Change the return code for signal frame related failures from explicit
error codes to a boolean fail/success as that's all what the calling
code evaluates.
- A large refactoring of the FPU code to prepare for adding AMX support:
- Distangle the public header maze and remove especially the misnomed
kitchen sink internal.h which is despite it's name included all over
the place.
- Add a proper abstraction for the register buffer storage (struct
fpstate) which allows to dynamically size the buffer at runtime by
flipping the pointer to the buffer container from the default
container which is embedded in task_struct::tread::fpu to a
dynamically allocated container with a larger register buffer.
- Convert the code over to the new fpstate mechanism.
- Consolidate the KVM FPU handling by moving the FPU related code into
the FPU core which removes the number of exports and avoids adding
even more export when AMX has to be supported in KVM. This also
removes duplicated code which was of course unnecessary different and
incomplete in the KVM copy.
- Simplify the KVM FPU buffer handling by utilizing the new fpstate
container and just switching the buffer pointer from the user space
buffer to the KVM guest buffer when entering vcpu_run() and flipping
it back when leaving the function. This cuts the memory requirements
of a vCPU for FPU buffers in half and avoids pointless memory copy
operations.
This also solves the so far unresolved problem of adding AMX support
because the current FPU buffer handling of KVM inflicted a circular
dependency between adding AMX support to the core and to KVM. With
the new scheme of switching fpstate AMX support can be added to the
core code without affecting KVM.
- Replace various variables with proper data structures so the extra
information required for adding dynamically enabled FPU features (AMX)
can be added in one place
- Add AMX (Advanved Matrix eXtensions) support (finally):
AMX is a large XSTATE component which is going to be available with
Saphire Rapids XEON CPUs. The feature comes with an extra MSR (MSR_XFD)
which allows to trap the (first) use of an AMX related instruction,
which has two benefits:
1) It allows the kernel to control access to the feature
2) It allows the kernel to dynamically allocate the large register
state buffer instead of burdening every task with the the extra 8K
or larger state storage.
It would have been great to gain this kind of control already with
AVX512.
The support comes with the following infrastructure components:
1) arch_prctl() to
- read the supported features (equivalent to XGETBV(0))
- read the permitted features for a task
- request permission for a dynamically enabled feature
Permission is granted per process, inherited on fork() and cleared
on exec(). The permission policy of the kernel is restricted to
sigaltstack size validation, but the syscall obviously allows
further restrictions via seccomp etc.
2) A stronger sigaltstack size validation for sys_sigaltstack(2) which
takes granted permissions and the potentially resulting larger
signal frame into account. This mechanism can also be used to
enforce factual sigaltstack validation independent of dynamic
features to help with finding potential victims of the 2K
sigaltstack size constant which is broken since AVX512 support was
added.
3) Exception handling for #NM traps to catch first use of a extended
feature via a new cause MSR. If the exception was caused by the use
of such a feature, the handler checks permission for that
feature. If permission has not been granted, the handler sends a
SIGILL like the #UD handler would do if the feature would have been
disabled in XCR0. If permission has been granted, then a new fpstate
which fits the larger buffer requirement is allocated.
In the unlikely case that this allocation fails, the handler sends
SIGSEGV to the task. That's not elegant, but unavoidable as the
other discussed options of preallocation or full per task
permissions come with their own set of horrors for kernel and/or
userspace. So this is the lesser of the evils and SIGSEGV caused by
unexpected memory allocation failures is not a fundamentally new
concept either.
When allocation succeeds, the fpstate properties are filled in to
reflect the extended feature set and the resulting sizes, the
fpu::fpstate pointer is updated accordingly and the trap is disarmed
for this task permanently.
4) Enumeration and size calculations
5) Trap switching via MSR_XFD
The XFD (eXtended Feature Disable) MSR is context switched with the
same life time rules as the FPU register state itself. The mechanism
is keyed off with a static key which is default disabled so !AMX
equipped CPUs have zero overhead. On AMX enabled CPUs the overhead
is limited by comparing the tasks XFD value with a per CPU shadow
variable to avoid redundant MSR writes. In case of switching from a
AMX using task to a non AMX using task or vice versa, the extra MSR
write is obviously inevitable.
All other places which need to be aware of the variable feature sets
and resulting variable sizes are not affected at all because they
retrieve the information (feature set, sizes) unconditonally from
the fpstate properties.
6) Enable the new AMX states
Note, this is relatively new code despite the fact that AMX support is in
the works for more than a year now.
The big refactoring of the FPU code, which allowed to do a proper
integration has been started exactly 3 weeks ago. Refactoring of the
existing FPU code and of the original AMX patches took a week and has
been subject to extensive review and testing. The only fallout which has
not been caught in review and testing right away was restricted to AMX
enabled systems, which is completely irrelevant for anyone outside Intel
and their early access program. There might be dragons lurking as usual,
but so far the fine grained refactoring has held up and eventual yet
undetected fallout is bisectable and should be easily addressable before
the 5.16 release. Famous last words...
Many thanks to Chang Bae and Dave Hansen for working hard on this and
also to the various test teams at Intel who reserved extra capacity to
follow the rapid development of this closely which provides the
confidence level required to offer this rather large update for inclusion
into 5.16-rc1.
-----BEGIN PGP SIGNATURE-----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=x3WL
-----END PGP SIGNATURE-----
Merge tag 'x86-fpu-2021-11-01' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 fpu updates from Thomas Gleixner:
- Cleanup of extable fixup handling to be more robust, which in turn
allows to make the FPU exception fixups more robust as well.
- Change the return code for signal frame related failures from
explicit error codes to a boolean fail/success as that's all what the
calling code evaluates.
- A large refactoring of the FPU code to prepare for adding AMX
support:
- Distangle the public header maze and remove especially the
misnomed kitchen sink internal.h which is despite it's name
included all over the place.
- Add a proper abstraction for the register buffer storage (struct
fpstate) which allows to dynamically size the buffer at runtime
by flipping the pointer to the buffer container from the default
container which is embedded in task_struct::tread::fpu to a
dynamically allocated container with a larger register buffer.
- Convert the code over to the new fpstate mechanism.
- Consolidate the KVM FPU handling by moving the FPU related code
into the FPU core which removes the number of exports and avoids
adding even more export when AMX has to be supported in KVM.
This also removes duplicated code which was of course
unnecessary different and incomplete in the KVM copy.
- Simplify the KVM FPU buffer handling by utilizing the new
fpstate container and just switching the buffer pointer from the
user space buffer to the KVM guest buffer when entering
vcpu_run() and flipping it back when leaving the function. This
cuts the memory requirements of a vCPU for FPU buffers in half
and avoids pointless memory copy operations.
This also solves the so far unresolved problem of adding AMX
support because the current FPU buffer handling of KVM inflicted
a circular dependency between adding AMX support to the core and
to KVM. With the new scheme of switching fpstate AMX support can
be added to the core code without affecting KVM.
- Replace various variables with proper data structures so the
extra information required for adding dynamically enabled FPU
features (AMX) can be added in one place
- Add AMX (Advanced Matrix eXtensions) support (finally):
AMX is a large XSTATE component which is going to be available with
Saphire Rapids XEON CPUs. The feature comes with an extra MSR
(MSR_XFD) which allows to trap the (first) use of an AMX related
instruction, which has two benefits:
1) It allows the kernel to control access to the feature
2) It allows the kernel to dynamically allocate the large register
state buffer instead of burdening every task with the the extra
8K or larger state storage.
It would have been great to gain this kind of control already with
AVX512.
The support comes with the following infrastructure components:
1) arch_prctl() to
- read the supported features (equivalent to XGETBV(0))
- read the permitted features for a task
- request permission for a dynamically enabled feature
Permission is granted per process, inherited on fork() and
cleared on exec(). The permission policy of the kernel is
restricted to sigaltstack size validation, but the syscall
obviously allows further restrictions via seccomp etc.
2) A stronger sigaltstack size validation for sys_sigaltstack(2)
which takes granted permissions and the potentially resulting
larger signal frame into account. This mechanism can also be used
to enforce factual sigaltstack validation independent of dynamic
features to help with finding potential victims of the 2K
sigaltstack size constant which is broken since AVX512 support
was added.
3) Exception handling for #NM traps to catch first use of a extended
feature via a new cause MSR. If the exception was caused by the
use of such a feature, the handler checks permission for that
feature. If permission has not been granted, the handler sends a
SIGILL like the #UD handler would do if the feature would have
been disabled in XCR0. If permission has been granted, then a new
fpstate which fits the larger buffer requirement is allocated.
In the unlikely case that this allocation fails, the handler
sends SIGSEGV to the task. That's not elegant, but unavoidable as
the other discussed options of preallocation or full per task
permissions come with their own set of horrors for kernel and/or
userspace. So this is the lesser of the evils and SIGSEGV caused
by unexpected memory allocation failures is not a fundamentally
new concept either.
When allocation succeeds, the fpstate properties are filled in to
reflect the extended feature set and the resulting sizes, the
fpu::fpstate pointer is updated accordingly and the trap is
disarmed for this task permanently.
4) Enumeration and size calculations
5) Trap switching via MSR_XFD
The XFD (eXtended Feature Disable) MSR is context switched with
the same life time rules as the FPU register state itself. The
mechanism is keyed off with a static key which is default
disabled so !AMX equipped CPUs have zero overhead. On AMX enabled
CPUs the overhead is limited by comparing the tasks XFD value
with a per CPU shadow variable to avoid redundant MSR writes. In
case of switching from a AMX using task to a non AMX using task
or vice versa, the extra MSR write is obviously inevitable.
All other places which need to be aware of the variable feature
sets and resulting variable sizes are not affected at all because
they retrieve the information (feature set, sizes) unconditonally
from the fpstate properties.
6) Enable the new AMX states
Note, this is relatively new code despite the fact that AMX support
is in the works for more than a year now.
The big refactoring of the FPU code, which allowed to do a proper
integration has been started exactly 3 weeks ago. Refactoring of the
existing FPU code and of the original AMX patches took a week and has
been subject to extensive review and testing. The only fallout which
has not been caught in review and testing right away was restricted
to AMX enabled systems, which is completely irrelevant for anyone
outside Intel and their early access program. There might be dragons
lurking as usual, but so far the fine grained refactoring has held up
and eventual yet undetected fallout is bisectable and should be
easily addressable before the 5.16 release. Famous last words...
Many thanks to Chang Bae and Dave Hansen for working hard on this and
also to the various test teams at Intel who reserved extra capacity
to follow the rapid development of this closely which provides the
confidence level required to offer this rather large update for
inclusion into 5.16-rc1
* tag 'x86-fpu-2021-11-01' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (110 commits)
Documentation/x86: Add documentation for using dynamic XSTATE features
x86/fpu: Include vmalloc.h for vzalloc()
selftests/x86/amx: Add context switch test
selftests/x86/amx: Add test cases for AMX state management
x86/fpu/amx: Enable the AMX feature in 64-bit mode
x86/fpu: Add XFD handling for dynamic states
x86/fpu: Calculate the default sizes independently
x86/fpu/amx: Define AMX state components and have it used for boot-time checks
x86/fpu/xstate: Prepare XSAVE feature table for gaps in state component numbers
x86/fpu/xstate: Add fpstate_realloc()/free()
x86/fpu/xstate: Add XFD #NM handler
x86/fpu: Update XFD state where required
x86/fpu: Add sanity checks for XFD
x86/fpu: Add XFD state to fpstate
x86/msr-index: Add MSRs for XFD
x86/cpufeatures: Add eXtended Feature Disabling (XFD) feature bit
x86/fpu: Reset permission and fpstate on exec()
x86/fpu: Prepare fpu_clone() for dynamically enabled features
x86/fpu/signal: Prepare for variable sigframe length
x86/signal: Use fpu::__state_user_size for sigalt stack validation
...
PKRU code does not need anything from FPU headers. Include cpufeature.h
instead and fixup the resulting fallout in perf.
This is a preparation for FPU changes in order to prevent recursive include
hell.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Borislav Petkov <bp@suse.de>
Link: https://lkml.kernel.org/r/20211015011538.551522694@linutronix.de
PEBS-via-PT records contain a mask of applicable counters. To identify
which event belongs to which counter, a side-band event is needed. Until
now, there has been no side-band event, and consequently users were limited
to using a single event.
Add such a side-band event. Note the event is optimised to output only
when the counter index changes for an event. That works only so long as
all PEBS-via-PT events are scheduled together, which they are for a
recording session because they are in a single group.
Also no attribute bit is used to select the new event, so a new
kernel is not compatible with older perf tools. The assumption
being that PEBS-via-PT is sufficiently esoteric that users will not
be troubled by this.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20210907163903.11820-2-adrian.hunter@intel.com
The typical way to access branch record (e.g. Intel LBR) is via hardware
perf_event. For CPUs with FREEZE_LBRS_ON_PMI support, PMI could capture
reliable LBR. On the other hand, LBR could also be useful in non-PMI
scenario. For example, in kretprobe or bpf fexit program, LBR could
provide a lot of information on what happened with the function. Add API
to use branch record for software use.
Note that, when the software event triggers, it is necessary to stop the
branch record hardware asap. Therefore, static_call is used to remove some
branch instructions in this process.
Suggested-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Acked-by: Andrii Nakryiko <andrii@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/bpf/20210910183352.3151445-2-songliubraving@fb.com
A warning as below may be occasionally triggered in an ADL machine when
these conditions occur:
- Two perf record commands run one by one. Both record a PEBS event.
- Both runs on small cores.
- They have different adaptive PEBS configuration (PEBS_DATA_CFG).
[ ] WARNING: CPU: 4 PID: 9874 at arch/x86/events/intel/ds.c:1743 setup_pebs_adaptive_sample_data+0x55e/0x5b0
[ ] RIP: 0010:setup_pebs_adaptive_sample_data+0x55e/0x5b0
[ ] Call Trace:
[ ] <NMI>
[ ] intel_pmu_drain_pebs_icl+0x48b/0x810
[ ] perf_event_nmi_handler+0x41/0x80
[ ] </NMI>
[ ] __perf_event_task_sched_in+0x2c2/0x3a0
Different from the big core, the small core requires the ACK right
before re-enabling counters in the NMI handler, otherwise a stale PEBS
record may be dumped into the later NMI handler, which trigger the
warning.
Add a new mid_ack flag to track the case. Add all PMI handler bits in
the struct x86_hybrid_pmu to track the bits for different types of
PMUs. Apply mid ACK for the small cores on an Alder Lake machine.
The existing hybrid() macro has a compile error when taking address of
a bit-field variable. Add a new macro hybrid_bit() to get the
bit-field value of a given PMU.
Fixes: f83d2f91d2 ("perf/x86/intel: Add Alder Lake Hybrid support")
Reported-by: Ammy Yi <ammy.yi@intel.com>
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Andi Kleen <ak@linux.intel.com>
Tested-by: Ammy Yi <ammy.yi@intel.com>
Link: https://lkml.kernel.org/r/1627997128-57891-1-git-send-email-kan.liang@linux.intel.com
If we use "perf record" in an AMD Milan guest, dmesg reports a #GP
warning from an unchecked MSR access error on MSR_F15H_PERF_CTLx:
[] unchecked MSR access error: WRMSR to 0xc0010200 (tried to write 0x0000020000110076) at rIP: 0xffffffff8106ddb4 (native_write_msr+0x4/0x20)
[] Call Trace:
[] amd_pmu_disable_event+0x22/0x90
[] x86_pmu_stop+0x4c/0xa0
[] x86_pmu_del+0x3a/0x140
The AMD64_EVENTSEL_HOSTONLY bit is defined and used on the host,
while the guest perf driver should avoid such use.
Fixes: 1018faa6cf ("perf/x86/kvm: Fix Host-Only/Guest-Only counting with SVM disabled")
Signed-off-by: Like Xu <likexu@tencent.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Liam Merwick <liam.merwick@oracle.com>
Tested-by: Kim Phillips <kim.phillips@amd.com>
Tested-by: Liam Merwick <liam.merwick@oracle.com>
Link: https://lkml.kernel.org/r/20210802070850.35295-1-likexu@tencent.com
- Platform PMU driver updates:
- x86 Intel uncore driver updates for Skylake (SNR) and Icelake (ICX) servers
- Fix RDPMC support
- Fix [extended-]PEBS-via-PT support
- Fix Sapphire Rapids event constraints
- Fix :ppp support on Sapphire Rapids
- Fix fixed counter sanity check on Alder Lake & X86_FEATURE_HYBRID_CPU
- Other heterogenous-PMU fixes
- Kprobes:
- Remove the unused and misguided kprobe::fault_handler callbacks.
- Warn about kprobes taking a page fault.
- Fix the 'nmissed' stat counter.
- Misc cleanups and fixes.
Signed-off-by: Ingo Molnar <mingo@kernel.org>
-----BEGIN PGP SIGNATURE-----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=20Vc
-----END PGP SIGNATURE-----
Merge tag 'perf-core-2021-06-28' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull perf events updates from Ingo Molnar:
- Platform PMU driver updates:
- x86 Intel uncore driver updates for Skylake (SNR) and Icelake (ICX) servers
- Fix RDPMC support
- Fix [extended-]PEBS-via-PT support
- Fix Sapphire Rapids event constraints
- Fix :ppp support on Sapphire Rapids
- Fix fixed counter sanity check on Alder Lake & X86_FEATURE_HYBRID_CPU
- Other heterogenous-PMU fixes
- Kprobes:
- Remove the unused and misguided kprobe::fault_handler callbacks.
- Warn about kprobes taking a page fault.
- Fix the 'nmissed' stat counter.
- Misc cleanups and fixes.
* tag 'perf-core-2021-06-28' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
perf: Fix task context PMU for Hetero
perf/x86/intel: Fix instructions:ppp support in Sapphire Rapids
perf/x86/intel: Add more events requires FRONTEND MSR on Sapphire Rapids
perf/x86/intel: Fix fixed counter check warning for some Alder Lake
perf/x86/intel: Fix PEBS-via-PT reload base value for Extended PEBS
perf/x86: Reset the dirty counter to prevent the leak for an RDPMC task
kprobes: Do not increment probe miss count in the fault handler
x86,kprobes: WARN if kprobes tries to handle a fault
kprobes: Remove kprobe::fault_handler
uprobes: Update uprobe_write_opcode() kernel-doc comment
perf/hw_breakpoint: Fix DocBook warnings in perf hw_breakpoint
perf/core: Fix DocBook warnings
perf/core: Make local function perf_pmu_snapshot_aux() static
perf/x86/intel/uncore: Enable I/O stacks to IIO PMON mapping on ICX
perf/x86/intel/uncore: Enable I/O stacks to IIO PMON mapping on SNR
perf/x86/intel/uncore: Generalize I/O stacks to PMON mapping procedure
perf/x86/intel/uncore: Drop unnecessary NULL checks after container_of()
The counter value of a perf task may leak to another RDPMC task.
For example, a perf stat task as below is running on CPU 0.
perf stat -e 'branches,cycles' -- taskset -c 0 ./workload
In the meantime, an RDPMC task, which is also running on CPU 0, may read
the GP counters periodically. (The RDPMC task creates a fixed event,
but read four GP counters.)
$./rdpmc_read_all_counters
index 0x0 value 0x8001e5970f99
index 0x1 value 0x8005d750edb6
index 0x2 value 0x0
index 0x3 value 0x0
index 0x0 value 0x8002358e48a5
index 0x1 value 0x8006bd1e3bc9
index 0x2 value 0x0
index 0x3 value 0x0
It is a potential security issue. Once the attacker knows what the other
thread is counting. The PerfMon counter can be used as a side-channel to
attack cryptosystems.
The counter value of the perf stat task leaks to the RDPMC task because
perf never clears the counter when it's stopped.
Three methods were considered to address the issue.
- Unconditionally reset the counter in x86_pmu_del(). It can bring extra
overhead even when there is no RDPMC task running.
- Only reset the un-assigned dirty counters when the RDPMC task is
scheduled in via sched_task(). It fails for the below case.
Thread A Thread B
clone(CLONE_THREAD) --->
set_affine(0)
set_affine(1)
while (!event-enabled)
;
event = perf_event_open()
mmap(event)
ioctl(event, IOC_ENABLE); --->
RDPMC
Counters are still leaked to the thread B.
- Only reset the un-assigned dirty counters before updating the CR4.PCE
bit. The method is implemented here.
The dirty counter is a counter, on which the assigned event has been
deleted, but the counter is not reset. To track the dirty counters,
add a 'dirty' variable in the struct cpu_hw_events.
The security issue can only be found with an RDPMC task. To enable the
RDMPC, the CR4.PCE bit has to be updated. Add a
perf_clear_dirty_counters() right before updating the CR4.PCE bit to
clear the existing dirty counters. Only the current un-assigned dirty
counters are reset, because the RDPMC assigned dirty counters will be
updated soon.
After applying the patch,
$ ./rdpmc_read_all_counters
index 0x0 value 0x0
index 0x1 value 0x0
index 0x2 value 0x0
index 0x3 value 0x0
index 0x0 value 0x0
index 0x1 value 0x0
index 0x2 value 0x0
index 0x3 value 0x0
Performance
The performance of a context switch only be impacted when there are two
or more perf users and one of the users must be an RDPMC user. In other
cases, there is no performance impact.
The worst-case occurs when there are two users: the RDPMC user only
uses one counter; while the other user uses all available counters.
When the RDPMC task is scheduled in, all the counters, other than the
RDPMC assigned one, have to be reset.
Test results for the worst-case, using a modified lat_ctx as measured
on an Ice Lake platform, which has 8 GP and 3 FP counters (ignoring
SLOTS).
lat_ctx -s 128K -N 1000 processes 2
Without the patch:
The context switch time is 4.97 us
With the patch:
The context switch time is 5.16 us
There is ~4% performance drop for the context switching time in the
worst-case.
Suggested-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/1623693582-187370-1-git-send-email-kan.liang@linux.intel.com
If the kernel is compiled with the CONFIG_LOCKDEP option, the conditional
might_sleep_if() deep in kmem_cache_alloc() will generate the following
trace, and potentially cause a deadlock when another LBR event is added:
[] BUG: sleeping function called from invalid context at include/linux/sched/mm.h:196
[] Call Trace:
[] kmem_cache_alloc+0x36/0x250
[] intel_pmu_lbr_add+0x152/0x170
[] x86_pmu_add+0x83/0xd0
Make it symmetric with the release_lbr_buffers() call and mirror the
existing DS buffers.
Fixes: c085fb8774 ("perf/x86/intel/lbr: Support XSAVES for arch LBR read")
Signed-off-by: Like Xu <like.xu@linux.intel.com>
[peterz: simplified]
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Kan Liang <kan.liang@linux.intel.com>
Link: https://lkml.kernel.org/r/20210430052247.3079672-2-like.xu@linux.intel.com
Alder Lake Hybrid system has two different types of core, Golden Cove
core and Gracemont core. The Golden Cove core is registered to
"cpu_core" PMU. The Gracemont core is registered to "cpu_atom" PMU.
The difference between the two PMUs include:
- Number of GP and fixed counters
- Events
- The "cpu_core" PMU supports Topdown metrics.
The "cpu_atom" PMU supports PEBS-via-PT.
The "cpu_core" PMU is similar to the Sapphire Rapids PMU, but without
PMEM.
The "cpu_atom" PMU is similar to Tremont, but with different events,
event_constraints, extra_regs and number of counters.
The mem-loads AUX event workaround only applies to the Golden Cove core.
Users may disable all CPUs of the same CPU type on the command line or
in the BIOS. For this case, perf still register a PMU for the CPU type
but the CPU mask is 0.
Current caps/pmu_name is usually the microarch codename. Assign the
"alderlake_hybrid" to the caps/pmu_name of both PMUs to indicate the
hybrid Alder Lake microarchitecture.
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Andi Kleen <ak@linux.intel.com>
Link: https://lkml.kernel.org/r/1618237865-33448-21-git-send-email-kan.liang@linux.intel.com
Implement filter_match callback for X86, which check whether an event is
schedulable on the current CPU.
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Andi Kleen <ak@linux.intel.com>
Link: https://lkml.kernel.org/r/1618237865-33448-20-git-send-email-kan.liang@linux.intel.com
Hybrid PMUs have different events and formats. In theory, Hybrid PMU
specific attributes should be maintained in the dedicated struct
x86_hybrid_pmu, but it wastes space because the events and formats are
similar among Hybrid PMUs.
To reduce duplication, all hybrid PMUs will share a group of attributes
in the following patch. To distinguish an attribute from different
Hybrid PMUs, a PMU aware attribute structure is introduced. A PMU type
is required for the attribute structure. The type is internal usage. It
is not visible in the sysfs API.
Hybrid PMUs may support the same event name, but with different event
encoding, e.g., the mem-loads event on an Atom PMU has different event
encoding from a Core PMU. It brings issue if two attributes are
created for them. Current sysfs_update_group finds an attribute by
searching the attr name (aka event name). If two attributes have the
same event name, the first attribute will be replaced.
To address the issue, only one attribute is created for the event. The
event_str is extended and stores event encodings from all Hybrid PMUs.
Each event encoding is divided by ";". The order of the event encodings
must follow the order of the hybrid PMU index. The event_str is internal
usage as well. When a user wants to show the attribute of a Hybrid PMU,
only the corresponding part of the string is displayed.
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Andi Kleen <ak@linux.intel.com>
Link: https://lkml.kernel.org/r/1618237865-33448-18-git-send-email-kan.liang@linux.intel.com
Different hybrid PMUs have different PMU capabilities and events. Perf
should registers a dedicated PMU for each of them.
To check the X86 event, perf has to go through all possible hybrid pmus.
All the hybrid PMUs are registered at boot time. Before the
registration, add intel_pmu_check_hybrid_pmus() to check and update the
counters information, the event constraints, the extra registers and the
unique capabilities for each hybrid PMUs.
Postpone the display of the PMU information and HW check to
CPU_STARTING, because the boot CPU is the only online CPU in the
init_hw_perf_events(). Perf doesn't know the availability of the other
PMUs. Perf should display the PMU information only if the counters of
the PMU are available.
One type of CPUs may be all offline. For this case, users can still
observe the PMU in /sys/devices, but its CPU mask is 0.
All hybrid PMUs have capability PERF_PMU_CAP_HETEROGENEOUS_CPUS.
The PMU name for hybrid PMUs will be "cpu_XXX", which will be assigned
later in a separated patch.
The PMU type id for the core PMU is still PERF_TYPE_RAW. For the other
hybrid PMUs, the PMU type id is not hard code.
The event->cpu must be compatitable with the supported CPUs of the PMU.
Add a check in the x86_pmu_event_init().
The events in a group must be from the same type of hybrid PMU.
The fake cpuc used in the validation must be from the supported CPU of
the event->pmu.
Perf may not retrieve a valid core type from get_this_hybrid_cpu_type().
For example, ADL may have an alternative configuration. With that
configuration, Perf cannot retrieve the core type from the CPUID leaf
0x1a. Add a platform specific get_hybrid_cpu_type(). If the generic way
fails, invoke the platform specific get_hybrid_cpu_type().
Suggested-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/1618237865-33448-17-git-send-email-kan.liang@linux.intel.com
The PMU capabilities are different among hybrid PMUs. Perf should dump
the PMU capabilities information for each hybrid PMU.
Factor out x86_pmu_show_pmu_cap() which shows the PMU capabilities
information. The function will be reused later when registering a
dedicated hybrid PMU.
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Andi Kleen <ak@linux.intel.com>
Link: https://lkml.kernel.org/r/1618237865-33448-16-git-send-email-kan.liang@linux.intel.com
Different hybrid PMU may have different extra registers, e.g. Core PMU
may have offcore registers, frontend register and ldlat register. Atom
core may only have offcore registers and ldlat register. Each hybrid PMU
should use its own extra_regs.
An Intel Hybrid system should always have extra registers.
Unconditionally allocate shared_regs for Intel Hybrid system.
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Andi Kleen <ak@linux.intel.com>
Link: https://lkml.kernel.org/r/1618237865-33448-11-git-send-email-kan.liang@linux.intel.com
The events are different among hybrid PMUs. Each hybrid PMU should use
its own event constraints.
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Andi Kleen <ak@linux.intel.com>
Link: https://lkml.kernel.org/r/1618237865-33448-10-git-send-email-kan.liang@linux.intel.com
The hardware cache events are different among hybrid PMUs. Each hybrid
PMU should have its own hw cache event table.
Suggested-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/1618237865-33448-9-git-send-email-kan.liang@linux.intel.com
The unconstrained value depends on the number of GP and fixed counters.
Each hybrid PMU should use its own unconstrained.
Suggested-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/1618237865-33448-8-git-send-email-kan.liang@linux.intel.com
The number of GP and fixed counters are different among hybrid PMUs.
Each hybrid PMU should use its own counter related information.
When handling a certain hybrid PMU, apply the number of counters from
the corresponding hybrid PMU.
When reserving the counters in the initialization of a new event,
reserve all possible counters.
The number of counter recored in the global x86_pmu is for the
architecture counters which are available for all hybrid PMUs. KVM
doesn't support the hybrid PMU yet. Return the number of the
architecture counters for now.
For the functions only available for the old platforms, e.g.,
intel_pmu_drain_pebs_nhm(), nothing is changed.
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Andi Kleen <ak@linux.intel.com>
Link: https://lkml.kernel.org/r/1618237865-33448-7-git-send-email-kan.liang@linux.intel.com
The intel_ctrl is the counter mask of a PMU. The PMU counter information
may be different among hybrid PMUs, each hybrid PMU should use its own
intel_ctrl to check and access the counters.
When handling a certain hybrid PMU, apply the intel_ctrl from the
corresponding hybrid PMU.
When checking the HW existence, apply the PMU and number of counters
from the corresponding hybrid PMU as well. Perf will check the HW
existence for each Hybrid PMU before registration. Expose the
check_hw_exists() for a later patch.
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Andi Kleen <ak@linux.intel.com>
Link: https://lkml.kernel.org/r/1618237865-33448-6-git-send-email-kan.liang@linux.intel.com
Some platforms, e.g. Alder Lake, have hybrid architecture. Although most
PMU capabilities are the same, there are still some unique PMU
capabilities for different hybrid PMUs. Perf should register a dedicated
pmu for each hybrid PMU.
Add a new struct x86_hybrid_pmu, which saves the dedicated pmu and
capabilities for each hybrid PMU.
The architecture MSR, MSR_IA32_PERF_CAPABILITIES, only indicates the
architecture features which are available on all hybrid PMUs. The
architecture features are stored in the global x86_pmu.intel_cap.
For Alder Lake, the model-specific features are perf metrics and
PEBS-via-PT. The corresponding bits of the global x86_pmu.intel_cap
should be 0 for these two features. Perf should not use the global
intel_cap to check the features on a hybrid system.
Add a dedicated intel_cap in the x86_hybrid_pmu to store the
model-specific capabilities. Use the dedicated intel_cap to replace
the global intel_cap for thse two features. The dedicated intel_cap
will be set in the following "Add Alder Lake Hybrid support" patch.
Add is_hybrid() to distinguish a hybrid system. ADL may have an
alternative configuration. With that configuration, the
X86_FEATURE_HYBRID_CPU is not set. Perf cannot rely on the feature bit.
Add a new static_key_false, perf_is_hybrid, to indicate a hybrid system.
It will be assigned in the following "Add Alder Lake Hybrid support"
patch as well.
Suggested-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/1618237865-33448-5-git-send-email-kan.liang@linux.intel.com
Some platforms, e.g. Alder Lake, have hybrid architecture. In the same
package, there may be more than one type of CPU. The PMU capabilities
are different among different types of CPU. Perf will register a
dedicated PMU for each type of CPU.
Add a 'pmu' variable in the struct cpu_hw_events to track the dedicated
PMU of the current CPU.
Current x86_get_pmu() use the global 'pmu', which will be broken on a
hybrid platform. Modify it to apply the 'pmu' of the specific CPU.
Initialize the per-CPU 'pmu' variable with the global 'pmu'. There is
nothing changed for the non-hybrid platforms.
The is_x86_event() will be updated in the later patch ("perf/x86:
Register hybrid PMUs") for hybrid platforms. For the non-hybrid
platforms, nothing is changed here.
Suggested-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/1618237865-33448-4-git-send-email-kan.liang@linux.intel.com
The 'running' variable is only used in the P4 PMU. Current perf sets the
variable in the critical function x86_pmu_start(), which wastes cycles
for everybody not running on P4.
Move cpuc->running into the P4 specific p4_pmu_enable_event().
Add a static per-CPU 'p4_running' variable to replace the 'running'
variable in the struct cpu_hw_events. Saves space for the generic
structure.
The p4_pmu_enable_all() also invokes the p4_pmu_enable_event(), but it
should not set cpuc->running. Factor out __p4_pmu_enable_event() for
p4_pmu_enable_all().
Suggested-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/1618410990-21383-1-git-send-email-kan.liang@linux.intel.com
With Architectural Performance Monitoring Version 5, CPUID 10.ECX cpu
leaf indicates the fixed counter enumeration. This extends the previous
count to a bitmap which allows disabling even lower fixed counters.
It could be used by a Hypervisor.
The existing intel_ctrl variable is used to remember the bitmask of the
counters. All code that reads all counters is fixed to check this extra
bitmask.
Suggested-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Originally-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/1611873611-156687-6-git-send-email-kan.liang@linux.intel.com
Add perf core PMU support for the Intel Sapphire Rapids server, which is
the successor of the Intel Ice Lake server. The enabling code is based
on Ice Lake, but there are several new features introduced.
The event encoding is changed and simplified, e.g., the event codes
which are below 0x90 are restricted to counters 0-3. The event codes
which above 0x90 are likely to have no restrictions. The event
constraints, extra_regs(), and hardware cache events table are changed
accordingly.
A new Precise Distribution (PDist) facility is introduced, which
further minimizes the skid when a precise event is programmed on the GP
counter 0. Enable the Precise Distribution (PDist) facility with :ppp
event. For this facility to work, the period must be initialized with a
value larger than 127. Add spr_limit_period() to apply the limit for
:ppp event.
Two new data source fields, data block & address block, are added in the
PEBS Memory Info Record for the load latency event. To enable the
feature,
- An auxiliary event has to be enabled together with the load latency
event on Sapphire Rapids. A new flag PMU_FL_MEM_LOADS_AUX is
introduced to indicate the case. A new event, mem-loads-aux, is
exposed to sysfs for the user tool.
Add a check in hw_config(). If the auxiliary event is not detected,
return an unique error -ENODATA.
- The union perf_mem_data_src is extended to support the new fields.
- Ice Lake and earlier models do not support block information, but the
fields may be set by HW on some machines. Add pebs_no_block to
explicitly indicate the previous platforms which don't support the new
block fields. Accessing the new block fields are ignored on those
platforms.
A new store Latency facility is introduced, which leverages the PEBS
facility where it can provide additional information about sampled
stores. The additional information includes the data address, memory
auxiliary info (e.g. Data Source, STLB miss) and the latency of the
store access. To enable the facility, the new event (0x02cd) has to be
programed on the GP counter 0. A new flag PERF_X86_EVENT_PEBS_STLAT is
introduced to indicate the event. The store_latency_data() is introduced
to parse the memory auxiliary info.
The layout of access latency field of PEBS Memory Info Record has been
changed. Two latency, instruction latency (bit 15:0) and cache access
latency (bit 47:32) are recorded.
- The cache access latency is similar to previous memory access latency.
For loads, the latency starts by the actual cache access until the
data is returned by the memory subsystem.
For stores, the latency starts when the demand write accesses the L1
data cache and lasts until the cacheline write is completed in the
memory subsystem.
The cache access latency is stored in low 32bits of the sample type
PERF_SAMPLE_WEIGHT_STRUCT.
- The instruction latency starts by the dispatch of the load operation
for execution and lasts until completion of the instruction it belongs
to.
Add a new flag PMU_FL_INSTR_LATENCY to indicate the instruction
latency support. The instruction latency is stored in the bit 47:32
of the sample type PERF_SAMPLE_WEIGHT_STRUCT.
Extends the PERF_METRICS MSR to feature TMA method level 2 metrics. The
lower half of the register is the TMA level 1 metrics (legacy). The
upper half is also divided into four 8-bit fields for the new level 2
metrics. Expose all eight Topdown metrics events to user space.
The full description for the SPR features can be found at Intel
Architecture Instruction Set Extensions and Future Features
Programming Reference, 319433-041.
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/1611873611-156687-5-git-send-email-kan.liang@linux.intel.com
Intel Sapphire Rapids server will introduce 8 metrics events. Intel
Ice Lake only supports 4 metrics events. A perf tool user may mistakenly
use the unsupported events via RAW format on Ice Lake. The user can
still get a value from the unsupported Topdown metrics event once the
following Sapphire Rapids enabling patch is applied.
To enable the 8 metrics events on Intel Sapphire Rapids, the
INTEL_TD_METRIC_MAX has to be updated, which impacts the
is_metric_event(). The is_metric_event() is a generic function.
On Ice Lake, the newly added SPR metrics events will be mistakenly
accepted as metric events on creation. At runtime, the unsupported
Topdown metrics events will be updated.
Add a variable num_topdown_events in x86_pmu to indicate the available
number of the Topdown metrics event on the platform. Apply the number
into is_metric_event(). Only the supported Topdown metrics events
should be created as metrics events.
Apply the num_topdown_events in icl_update_topdown_event() as well. The
function can be reused by the following patch.
Suggested-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/1611873611-156687-4-git-send-email-kan.liang@linux.intel.com
Perfmon-v4 counter freezing is fundamentally broken; remove this default
disabled code to make sure nobody uses it.
The feature is called Freeze-on-PMI in the SDM, and if it would do that,
there wouldn't actually be a problem, *however* it does something subtly
different. It globally disables the whole PMU when it raises the PMI,
not when the PMI hits.
This means there's a window between the PMI getting raised and the PMI
actually getting served where we loose events and this violates the
perf counter independence. That is, a counting event should not result
in a different event count when there is a sampling event co-scheduled.
This is known to break existing software (RR).
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Starting with Arch Perfmon v5, the anythread filter on generic counters may be
deprecated. The current kernel was exporting the any filter without checking.
On Icelake, it means you could do cpu/event=0x3c,any/ even though the filter
does not exist. This patch corrects the problem by relying on the CPUID 0xa leaf
function to determine if anythread is supported or not as described in the
Intel SDM Vol3b 18.2.5.1 AnyThread Deprecation section.
Signed-off-by: Stephane Eranian <eranian@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20201028194247.3160610-1-eranian@google.com
intel_pmu_drain_pebs_*() is typically called from handle_pmi_common(),
both have an on-stack struct perf_sample_data, which is *big*. Rewire
things so that drain_pebs() can use the one handle_pmi_common() has.
Reported-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20201030151955.054099690@infradead.org
When studying code layout, it is useful to capture the page size of the
sampled code address.
Add a new sample type for code page size.
The new sample type requires collecting the ip. The code page size can
be calculated from the NMI-safe perf_get_page_size().
For large PEBS, it's very unlikely that the mapping is gone for the
earlier PEBS records. Enable the feature for the large PEBS. The worst
case is that page-size '0' is returned.
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Stephane Eranian <eranian@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20201001135749.2804-5-kan.liang@linux.intel.com
When a group that has TopDown members is failed to be scheduled, any
later TopDown groups will not return valid values.
Here is an example.
A background perf that occupies all the GP counters and the fixed
counter 1.
$perf stat -e "{cycles,cycles,cycles,cycles,cycles,cycles,cycles,
cycles,cycles}:D" -a
A user monitors a TopDown group. It works well, because the fixed
counter 3 and the PERF_METRICS are available.
$perf stat -x, --topdown -- ./workload
retiring,bad speculation,frontend bound,backend bound,
18.0,16.1,40.4,25.5,
Then the user tries to monitor a group that has TopDown members.
Because of the cycles event, the group is failed to be scheduled.
$perf stat -x, -e '{slots,topdown-retiring,topdown-be-bound,
topdown-fe-bound,topdown-bad-spec,cycles}'
-- ./workload
<not counted>,,slots,0,0.00,,
<not counted>,,topdown-retiring,0,0.00,,
<not counted>,,topdown-be-bound,0,0.00,,
<not counted>,,topdown-fe-bound,0,0.00,,
<not counted>,,topdown-bad-spec,0,0.00,,
<not counted>,,cycles,0,0.00,,
The user tries to monitor a TopDown group again. It doesn't work anymore.
$perf stat -x, --topdown -- ./workload
,,,,,
In a txn, cancel_txn() is to truncate the event_list for a canceled
group and update the number of events added in this transaction.
However, the number of TopDown events added in this transaction is not
updated. The kernel will probably fail to add new Topdown events.
Fixes: 7b2c05a15d ("perf/x86/intel: Generic support for hardware TopDown metrics")
Reported-by: Andi Kleen <ak@linux.intel.com>
Reported-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Kan Liang <kan.liang@linux.intel.com>
Link: https://lkml.kernel.org/r/20201005082611.GH2628@hirez.programming.kicks-ass.net
Kan reported that n_metric gets corrupted for cancelled transactions;
a similar issue exists for n_pair for AMD's Large Increment thing.
The problem was confirmed and confirmed fixed by Kim using:
sudo perf stat -e "{cycles,cycles,cycles,cycles}:D" -a sleep 10 &
# should succeed:
sudo perf stat -e "{fp_ret_sse_avx_ops.all}:D" -a workload
# should fail:
sudo perf stat -e "{fp_ret_sse_avx_ops.all,fp_ret_sse_avx_ops.all,cycles}:D" -a workload
# previously failed, now succeeds with this patch:
sudo perf stat -e "{fp_ret_sse_avx_ops.all}:D" -a workload
Fixes: 5738891229 ("perf/x86/amd: Add support for Large Increment per Cycle Events")
Reported-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Tested-by: Kim Phillips <kim.phillips@amd.com>
Link: https://lkml.kernel.org/r/20201005082516.GG2628@hirez.programming.kicks-ass.net
Ice Lake supports the hardware TopDown metrics feature, which can free
up the scarce GP counters.
Update the event constraints for the metrics events. The metric counters
do not exist, which are mapped to a dummy offset. The sharing between
multiple users of the same metric without multiplexing is not allowed.
Implement set_topdown_event_period for Ice Lake. The values in
PERF_METRICS MSR are derived from the fixed counter 3. Both registers
should start from zero.
Implement update_topdown_event for Ice Lake. The metric is reported by
multiplying the metric (fraction) with slots. To maintain accurate
measurements, both registers are cleared for each update. The fixed
counter 3 should always be cleared before the PERF_METRICS.
Implement td_attr for the new metrics events and the new slots fixed
counter. Make them visible to the perf user tools.
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20200723171117.9918-11-kan.liang@linux.intel.com
Intro
=====
The TopDown Microarchitecture Analysis (TMA) Method is a structured
analysis methodology to identify critical performance bottlenecks in
out-of-order processors. Current perf has supported the method.
The method works well, but there is one problem. To collect the TopDown
events, several GP counters have to be used. If a user wants to collect
other events at the same time, the multiplexing probably be triggered,
which impacts the accuracy.
To free up the scarce GP counters, the hardware TopDown metrics feature
is introduced from Ice Lake. The hardware implements an additional
"metrics" register and a new Fixed Counter 3 that measures pipeline
"slots". The TopDown events can be calculated from them instead.
Events
======
The level 1 TopDown has four metrics. There is no event-code assigned to
the TopDown metrics. Four metric events are exported as separate perf
events, which map to the internal "metrics" counter register. Those
events do not exist in hardware, but can be allocated by the scheduler.
For the event mapping, a special 0x00 event code is used, which is
reserved for fake events. The metric events start from umask 0x10.
When setting up the metric events, they point to the Fixed Counter 3.
They have to be specially handled.
- Add the update_topdown_event() callback to read the additional metrics
MSR and generate the metrics.
- Add the set_topdown_event_period() callback to initialize metrics MSR
and the fixed counter 3.
- Add a variable n_metric_event to track the number of the accepted
metrics events. The sharing between multiple users of the same metric
without multiplexing is not allowed.
- Only enable/disable the fixed counter 3 when there are no other active
TopDown events, which avoid the unnecessary writing of the fixed
control register.
- Disable the PMU when reading the metrics event. The metrics MSR and
the fixed counter 3 are read separately. The values may be modified by
an NMI.
All four metric events don't support sampling. Since they will be
handled specially for event update, a flag PERF_X86_EVENT_TOPDOWN is
introduced to indicate this case.
The slots event can support both sampling and counting.
For counting, the flag is also applied.
For sampling, it will be handled normally as other normal events.
Groups
======
The slots event is required in a Topdown group.
To avoid reading the METRICS register multiple times, the metrics and
slots value can only be updated by slots event in a group.
All active slots and metrics events will be updated one time.
Therefore, the slots event must be before any metric events in a Topdown
group.
NMI
======
The METRICS related register may be overflow. The bit 48 of the STATUS
register will be set. If so, PERF_METRICS and Fixed counter 3 are
required to be reset. The patch also update all active slots and
metrics events in the NMI handler.
The update_topdown_event() has to read two registers separately. The
values may be modified by an NMI. PMU has to be disabled before calling
the function.
RDPMC
======
RDPMC is temporarily disabled. A later patch will enable it.
Suggested-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20200723171117.9918-9-kan.liang@linux.intel.com
Bit 15 of the PERF_CAPABILITIES MSR indicates that the perf METRICS
feature is supported. The perf METRICS is not a PEBS feature.
Rename pebs_metrics_available perf_metrics.
The bit is not used in the current code. It will be used in a later
patch.
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20200723171117.9918-6-kan.liang@linux.intel.com
Reading LBR registers in a perf NMI handler for a non-PEBS event
causes a high overhead because the number of LBR registers is huge.
To reduce the overhead, the XSAVES instruction should be used to replace
the LBR registers' reading method.
The XSAVES buffer used for LBR read has to be per-CPU because the NMI
handler invoked the lbr_read(). The existing task_ctx_data buffer
cannot be used which is per-task and only be allocated for the LBR call
stack mode. A new lbr_xsave pointer is introduced in the cpu_hw_events
as an XSAVES buffer for LBR read.
The XSAVES buffer should be allocated only when LBR is used by a
non-PEBS event on the CPU because the total size of the lbr_xsave is
not small (~1.4KB).
The XSAVES buffer is allocated when a non-PEBS event is added, but it
is lazily released in x86_release_hardware() when perf releases the
entire PMU hardware resource, because perf may frequently schedule the
event, e.g. high context switch. The lazy release method reduces the
overhead of frequently allocate/free the buffer.
If the lbr_xsave fails to be allocated, roll back to normal Arch LBR
lbr_read().
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Dave Hansen <dave.hansen@intel.com>
Link: https://lkml.kernel.org/r/1593780569-62993-24-git-send-email-kan.liang@linux.intel.com
In the LBR call stack mode, LBR information is used to reconstruct a
call stack. To get the complete call stack, perf has to save/restore
all LBR registers during a context switch. Due to a large number of the
LBR registers, this process causes a high CPU overhead. To reduce the
CPU overhead during a context switch, use the XSAVES/XRSTORS
instructions.
Every XSAVE area must follow a canonical format: the legacy region, an
XSAVE header and the extended region. Although the LBR information is
only kept in the extended region, a space for the legacy region and
XSAVE header is still required. Add a new dedicated structure for LBR
XSAVES support.
Before enabling XSAVES support, the size of the LBR state has to be
sanity checked, because:
- the size of the software structure is calculated from the max number
of the LBR depth, which is enumerated by the CPUID leaf for Arch LBR.
The size of the LBR state is enumerated by the CPUID leaf for XSAVE
support of Arch LBR. If the values from the two CPUID leaves are not
consistent, it may trigger a buffer overflow. For example, a hypervisor
may unconsciously set inconsistent values for the two emulated CPUID.
- unlike other state components, the size of an LBR state depends on the
max number of LBRs, which may vary from generation to generation.
Expose the function xfeature_size() for the sanity check.
The LBR XSAVES support will be disabled if the size of the LBR state
enumerated by CPUID doesn't match with the size of the software
structure.
The XSAVE instruction requires 64-byte alignment for state buffers. A
new macro is added to reflect the alignment requirement. A 64-byte
aligned kmem_cache is created for architecture LBR.
Currently, the structure for each state component is maintained in
fpu/types.h. The structure for the new LBR state component should be
maintained in the same place. Move structure lbr_entry to fpu/types.h as
well for broader sharing.
Add dedicated lbr_save/lbr_restore functions for LBR XSAVES support,
which invokes the corresponding xstate helpers to XSAVES/XRSTORS LBR
information at the context switch when the call stack mode is enabled.
Since the XSAVES/XRSTORS instructions will be eventually invoked, the
dedicated functions is named with '_xsaves'/'_xrstors' postfix.
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Dave Hansen <dave.hansen@intel.com>
Link: https://lkml.kernel.org/r/1593780569-62993-23-git-send-email-kan.liang@linux.intel.com
Last Branch Records (LBR) enables recording of software path history by
logging taken branches and other control flows within architectural
registers now. Intel CPUs have had model-specific LBR for quite some
time, but this evolves them into an architectural feature now.
The main improvements of Architectural LBR implemented includes:
- Linux kernel can support the LBR features without knowing the model
number of the current CPU.
- Architectural LBR capabilities can be enumerated by CPUID. The
lbr_ctl_map is based on the CPUID Enumeration.
- The possible LBR depth can be retrieved from CPUID enumeration. The
max value is written to the new MSR_ARCH_LBR_DEPTH as the number of
LBR entries.
- A new IA32_LBR_CTL MSR is introduced to enable and configure LBRs,
which replaces the IA32_DEBUGCTL[bit 0] and the LBR_SELECT MSR.
- Each LBR record or entry is still comprised of three MSRs,
IA32_LBR_x_FROM_IP, IA32_LBR_x_TO_IP and IA32_LBR_x_TO_IP.
But they become the architectural MSRs.
- Architectural LBR is stack-like now. Entry 0 is always the youngest
branch, entry 1 the next youngest... The TOS MSR has been removed.
The way to enable/disable Architectural LBR is similar to the previous
model-specific LBR. __intel_pmu_lbr_enable/disable() can be reused, but
some modifications are required, which include:
- MSR_ARCH_LBR_CTL is used to enable and configure the Architectural
LBR.
- When checking the value of the IA32_DEBUGCTL MSR, ignoring the
DEBUGCTLMSR_LBR (bit 0) for Architectural LBR, which has no meaning
and always return 0.
- The FREEZE_LBRS_ON_PMI has to be explicitly set/clear, because
MSR_IA32_DEBUGCTLMSR is not touched in __intel_pmu_lbr_disable() for
Architectural LBR.
- Only MSR_ARCH_LBR_CTL is cleared in __intel_pmu_lbr_disable() for
Architectural LBR.
Some Architectural LBR dedicated functions are implemented to
reset/read/save/restore LBR.
- For reset, writing to the ARCH_LBR_DEPTH MSR clears all Arch LBR
entries, which is a lot faster and can improve the context switch
latency.
- For read, the branch type information can be retrieved from
the MSR_ARCH_LBR_INFO_*. But it's not fully compatible due to
OTHER_BRANCH type. The software decoding is still required for the
OTHER_BRANCH case.
LBR records are stored in the age order as well. Reuse
intel_pmu_store_lbr(). Check the CPUID enumeration before accessing
the corresponding bits in LBR_INFO.
- For save/restore, applying the fast reset (writing ARCH_LBR_DEPTH).
Reading 'lbr_from' of entry 0 instead of the TOS MSR to check if the
LBR registers are reset in the deep C-state. If 'the deep C-state
reset' bit is not set in CPUID enumeration, ignoring the check.
XSAVE support for Architectural LBR will be implemented later.
The number of LBR entries cannot be hardcoded anymore, which should be
retrieved from CPUID enumeration. A new structure
x86_perf_task_context_arch_lbr is introduced for Architectural LBR.
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/1593780569-62993-15-git-send-email-kan.liang@linux.intel.com
The previous model-specific LBR and Architecture LBR (legacy way) use a
similar method to save/restore the LBR information, which directly
accesses the LBR registers. The codes which read/write a set of LBR
registers can be shared between them.
Factor out two functions which are used to read/write a set of LBR
registers.
Add lbr_info into structure x86_pmu, and use it to replace the hardcoded
LBR INFO MSR, because the LBR INFO MSR address of the previous
model-specific LBR is different from Architecture LBR. The MSR address
should be assigned at boot time. For now, only Sky Lake and later
platforms have the LBR INFO MSR.
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/1593780569-62993-13-git-send-email-kan.liang@linux.intel.com
Current LBR information in the structure x86_perf_task_context is stored
in a different format from the PEBS LBR record and Architecture LBR,
which prevents the sharing of the common codes.
Use the format of the PEBS LBR record as a unified format. Use a generic
name lbr_entry to replace pebs_lbr_entry.
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/1593780569-62993-11-git-send-email-kan.liang@linux.intel.com
An IA32_LBR_CTL is introduced for Architecture LBR to enable and config
LBR registers to replace the previous LBR_SELECT.
All the related members in struct cpu_hw_events and struct x86_pmu
have to be renamed.
Some new macros are added to reflect the layout of LBR_CTL.
The mapping from PERF_SAMPLE_BRANCH_* to the corresponding bits in
LBR_CTL MSR is saved in lbr_ctl_map now, which is not a const value.
The value relies on the CPUID enumeration.
For the previous model-specific LBR, most of the bits in LBR_SELECT
operate in the suppressed mode. For the bits in LBR_CTL, the polarity is
inverted.
For the previous model-specific LBR format 5 (LBR_FORMAT_INFO), if the
NO_CYCLES and NO_FLAGS type are set, the flag LBR_NO_INFO will be set to
avoid the unnecessary LBR_INFO MSR read. Although Architecture LBR also
has a dedicated LBR_INFO MSR, perf doesn't need to check and set the
flag LBR_NO_INFO. For Architecture LBR, XSAVES instruction will be used
as the default way to read the LBR MSRs all together. The overhead which
the flag tries to avoid doesn't exist anymore. Dropping the flag can
save the extra check for the flag in the lbr_read() later, and make the
code cleaner.
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/1593780569-62993-10-git-send-email-kan.liang@linux.intel.com
The LBR capabilities of Architecture LBR are retrieved from the CPUID
enumeration once at boot time. The capabilities have to be saved for
future usage.
Several new fields are added into structure x86_pmu to indicate the
capabilities. The fields will be used in the following patches.
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/1593780569-62993-9-git-send-email-kan.liang@linux.intel.com
The type of task_ctx is hardcoded as struct x86_perf_task_context,
which doesn't apply for Architecture LBR. For example, Architecture LBR
doesn't have the TOS MSR. The number of LBR entries is variable. A new
struct will be introduced for Architecture LBR. Perf has to determine
the type of task_ctx at run time.
The type of task_ctx pointer is changed to 'void *', which will be
determined at run time.
The generic LBR optimization can be shared between Architecture LBR and
model-specific LBR. Both need to access the structure for the generic
LBR optimization. A helper task_context_opt() is introduced to retrieve
the pointer of the structure at run time.
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/1593780569-62993-7-git-send-email-kan.liang@linux.intel.com
To reduce the overhead of a context switch with LBR enabled, some
generic optimizations were introduced, e.g. avoiding restore LBR if no
one else touched them. The generic optimizations can also be used by
Architecture LBR later. Currently, the fields for the generic
optimizations are part of structure x86_perf_task_context, which will be
deprecated by Architecture LBR. A new structure should be introduced
for the common fields of generic optimization, which can be shared
between Architecture LBR and model-specific LBR.
Both 'valid_lbrs' and 'tos' are also used by the generic optimizations,
but they are not moved into the new structure, because Architecture LBR
is stack-like. The 'valid_lbrs' which records the index of the valid LBR
is not required anymore. The TOS MSR will be removed.
LBR registers may be cleared in the deep Cstate. If so, the generic
optimizations should not be applied. Perf has to unconditionally
restore the LBR registers. A generic function is required to detect the
reset due to the deep Cstate. lbr_is_reset_in_cstate() is introduced.
Currently, for the model-specific LBR, the TOS MSR is used to detect the
reset. There will be another method introduced for Architecture LBR
later.
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/1593780569-62993-6-git-send-email-kan.liang@linux.intel.com
The MSRs of Architectural LBR are different from previous model-specific
LBR. Perf has to implement different functions to save and restore them.
The function pointers for LBR save and restore are introduced. Perf
should initialize the corresponding functions at boot time.
The generic optimizations, e.g. avoiding restore LBR if no one else
touched them, still apply for Architectural LBRs. The related codes are
not moved to model-specific functions.
Current model-specific LBR functions are set as default.
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/1593780569-62993-5-git-send-email-kan.liang@linux.intel.com
The method to read Architectural LBRs is different from previous
model-specific LBR. Perf has to implement a different function.
A function pointer for LBR read is introduced. Perf should initialize
the corresponding function at boot time, and avoid checking lbr_format
at run time.
The current 64-bit LBR read function is set as default.
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/1593780569-62993-4-git-send-email-kan.liang@linux.intel.com
The method to reset Architectural LBRs is different from previous
model-specific LBR. Perf has to implement a different function.
A function pointer is introduced for LBR reset. The enum of
LBR_FORMAT_* is also moved to perf_event.h. Perf should initialize the
corresponding functions at boot time, and avoid checking lbr_format at
run time.
The current 64-bit LBR reset function is set as default.
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/1593780569-62993-3-git-send-email-kan.liang@linux.intel.com
