| Age | Commit message (Collapse) | Author |
|---|
|
* for-next/perf-user-counter-access:
Documentation: arm64: Document PMU counters access from userspace
arm64: perf: Enable PMU counter userspace access for perf event
arm64: perf: Add userspace counter access disable switch
perf: Add a counter for number of user access events in context
x86: perf: Move RDPMC event flag to a common definition
|
|
* for-next/perf-smmu:
perf/smmuv3: Synthesize IIDR from CoreSight ID registers
perf/smmuv3: Add devicetree support
dt-bindings: Add Arm SMMUv3 PMCG binding
|
|
* for-next/perf-hisi:
drivers/perf: hisi: Add driver for HiSilicon PCIe PMU
docs: perf: Add description for HiSilicon PCIe PMU driver
|
|
* for-next/perf-cn10k:
dt-bindings: perf: Add YAML schemas for Marvell CN10K LLC-TAD pmu bindings
drivers: perf: Add LLC-TAD perf counter support
|
|
* for-next/perf-cmn:
perf/arm-cmn: Add debugfs topology info
perf/arm-cmn: Add CI-700 Support
dt-bindings: perf: arm-cmn: Add CI-700
perf/arm-cmn: Support new IP features
perf/arm-cmn: Demarcate CMN-600 specifics
perf/arm-cmn: Move group validation data off-stack
perf/arm-cmn: Optimise DTC counter accesses
perf/arm-cmn: Optimise DTM counter reads
perf/arm-cmn: Refactor DTM handling
perf/arm-cmn: Streamline node iteration
perf/arm-cmn: Refactor node ID handling
perf/arm-cmn: Drop compile-test restriction
perf/arm-cmn: Account for NUMA affinity
perf/arm-cmn: Fix CPU hotplug unregistration
|
|
Some powers were changed during the jack insert detection
and clk's enable/disable in CCF.
If in parallel, the influence has a chance to detect
the wrong jack type, so add a lock.
Signed-off-by: Derek Fang <derek.fang@realtek.com>
Link: https://lore.kernel.org/r/20211214105033.471-1-derek.fang@realtek.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
The FEAT_LSE atomic instructions include LD* instructions which return
the original value of a memory location can be used to directly
implement FETCH opertations. Each RETURN op is implemented as a copy of
the corresponding FETCH op with a trailing instruction to generate the
new value of the memory location. We only directly implement
*_fetch_add*(), for which we have a trailing `add` instruction.
As the compiler has no visibility of the `add`, this leads to less than
optimal code generation when consuming the result.
For example, the compiler cannot constant-fold the addition into later
operations, and currently GCC 11.1.0 will compile:
return __lse_atomic_sub_return(1, v) == 0;
As:
mov w1, #0xffffffff
ldaddal w1, w2, [x0]
add w1, w1, w2
cmp w1, #0x0
cset w0, eq // eq = none
ret
This patch improves this by replacing the `add` with C addition after
the inline assembly block, e.g.
ret += i;
This allows the compiler to manipulate `i`. This permits the compiler to
merge the `add` and `cmp` for the above, e.g.
mov w1, #0xffffffff
ldaddal w1, w1, [x0]
cmp w1, #0x1
cset w0, eq // eq = none
ret
With this change the assembly for each RETURN op is identical to the
corresponding FETCH op (including barriers and clobbers) so I've removed
the inline assembly and rewritten each RETURN op in terms of the
corresponding FETCH op, e.g.
| static inline void __lse_atomic_add_return(int i, atomic_t *v)
| {
| return __lse_atomic_fetch_add(i, v) + i
| }
The new construction does not adversely affect the common case, and
before and after this patch GCC 11.1.0 can compile:
__lse_atomic_add_return(i, v)
As:
ldaddal w0, w2, [x1]
add w0, w0, w2
... while having the freedom to do better elsewhere.
This is intended as an optimization and cleanup.
There should be no functional change as a result of this patch.
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Cc: Boqun Feng <boqun.feng@gmail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Will Deacon <will@kernel.org>
Acked-by: Will Deacon <will@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20211210151410.2782645-6-mark.rutland@arm.com
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
|
|
We have overly conservative assembly constraints for the basic FEAT_LSE
atomic instructions, and using more accurate and permissive constraints
will allow for better code generation.
The FEAT_LSE basic atomic instructions have come in two forms:
LD{op}{order}{size} <Rs>, <Rt>, [<Rn>]
ST{op}{order}{size} <Rs>, [<Rn>]
The ST* forms are aliases of the LD* forms where:
ST{op}{order}{size} <Rs>, [<Rn>]
Is:
LD{op}{order}{size} <Rs>, XZR, [<Rn>]
For either form, both <Rs> and <Rn> are read but not written back to,
and <Rt> is written with the original value of the memory location.
Where (<Rt> == <Rs>) or (<Rt> == <Rn>), <Rt> is written *after* the
other register value(s) are consumed. There are no UNPREDICTABLE or
CONSTRAINED UNPREDICTABLE behaviours when any pair of <Rs>, <Rt>, or
<Rn> are the same register.
Our current inline assembly always uses <Rs> == <Rt>, treating this
register as both an input and an output (using a '+r' constraint). This
forces the compiler to do some unnecessary register shuffling and/or
redundant value generation.
For example, the compiler cannot reuse the <Rs> value, and currently GCC
11.1.0 will compile:
__lse_atomic_add(1, a);
__lse_atomic_add(1, b);
__lse_atomic_add(1, c);
As:
mov w3, #0x1
mov w4, w3
stadd w4, [x0]
mov w0, w3
stadd w0, [x1]
stadd w3, [x2]
We can improve this with more accurate constraints, separating <Rs> and
<Rt>, where <Rs> is an input-only register ('r'), and <Rt> is an
output-only value ('=r'). As <Rt> is written back after <Rs> is
consumed, it does not need to be earlyclobber ('=&r'), leaving the
compiler free to use the same register for both <Rs> and <Rt> where this
is desirable.
At the same time, the redundant 'r' constraint for `v` is removed, as
the `+Q` constraint is sufficient.
With this change, the above example becomes:
mov w3, #0x1
stadd w3, [x0]
stadd w3, [x1]
stadd w3, [x2]
I've made this change for the non-value-returning and FETCH ops. The
RETURN ops have a multi-instruction sequence for which we cannot use the
same constraints, and a subsequent patch will rewrite hte RETURN ops in
terms of the FETCH ops, relying on the ability for the compiler to reuse
the <Rs> value.
This is intended as an optimization.
There should be no functional change as a result of this patch.
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Cc: Boqun Feng <boqun.feng@gmail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Will Deacon <will@kernel.org>
Acked-by: Will Deacon <will@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20211210151410.2782645-5-mark.rutland@arm.com
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
|
|
The FEAT_LSE atomic instructions include atomic bit-clear instructions
(`ldclr*` and `stclr*`) which can be used to directly implement ANDNOT
operations. Each AND op is implemented as a copy of the corresponding
ANDNOT op with a leading `mvn` instruction to apply a bitwise NOT to the
`i` argument.
As the compiler has no visibility of the `mvn`, this leads to less than
optimal code generation when generating `i` into a register. For
example, __lse_atomic_fetch_and(0xf, v) can be compiled to:
mov w1, #0xf
mvn w1, w1
ldclral w1, w1, [x2]
This patch improves this by replacing the `mvn` with NOT in C before the
inline assembly block, e.g.
i = ~i;
This allows the compiler to generate `i` into a register more optimally,
e.g.
mov w1, #0xfffffff0
ldclral w1, w1, [x2]
With this change the assembly for each AND op is identical to the
corresponding ANDNOT op (including barriers and clobbers), so I've
removed the inline assembly and rewritten each AND op in terms of the
corresponding ANDNOT op, e.g.
| static inline void __lse_atomic_and(int i, atomic_t *v)
| {
| return __lse_atomic_andnot(~i, v);
| }
This is intended as an optimization and cleanup.
There should be no functional change as a result of this patch.
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Cc: Boqun Feng <boqun.feng@gmail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Will Deacon <will@kernel.org>
Acked-by: Will Deacon <will@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20211210151410.2782645-4-mark.rutland@arm.com
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
|
|
The FEAT_LSE atomic instructions include atomic ADD instructions
(`stadd*` and `ldadd*`), but do not include atomic SUB instructions, so
we must build all of the SUB operations using the ADD instructions. We
open-code these today, with each SUB op implemented as a copy of the
corresponding ADD op with a leading `neg` instruction in the inline
assembly to negate the `i` argument.
As the compiler has no visibility of the `neg`, this leads to less than
optimal code generation when generating `i` into a register. For
example, __les_atomic_fetch_sub(1, v) can be compiled to:
mov w1, #0x1
neg w1, w1
ldaddal w1, w1, [x2]
This patch improves this by replacing the `neg` with negation in C
before the inline assembly block, e.g.
i = -i;
This allows the compiler to generate `i` into a register more optimally,
e.g.
mov w1, #0xffffffff
ldaddal w1, w1, [x2]
With this change the assembly for each SUB op is identical to the
corresponding ADD op (including barriers and clobbers), so I've removed
the inline assembly and rewritten each SUB op in terms of the
corresponding ADD op, e.g.
| static inline void __lse_atomic_sub(int i, atomic_t *v)
| {
| __lse_atomic_add(-i, v);
| }
For clarity I've moved the definition of each SUB op immediately after
the corresponding ADD op, and used a single macro to create the RETURN
forms of both ops.
This is intended as an optimization and cleanup.
There should be no functional change as a result of this patch.
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Cc: Boqun Feng <boqun.feng@gmail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Will Deacon <will@kernel.org>
Acked-by: Will Deacon <will@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20211210151410.2782645-3-mark.rutland@arm.com
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
|
|
The code for the atomic ops is formatted inconsistently, and while this
is not a functional problem it is rather distracting when working on
them.
Some have ops have consistent indentation, e.g.
| #define ATOMIC_OP_ADD_RETURN(name, mb, cl...) \
| static inline int __lse_atomic_add_return##name(int i, atomic_t *v) \
| { \
| u32 tmp; \
| \
| asm volatile( \
| __LSE_PREAMBLE \
| " ldadd" #mb " %w[i], %w[tmp], %[v]\n" \
| " add %w[i], %w[i], %w[tmp]" \
| : [i] "+r" (i), [v] "+Q" (v->counter), [tmp] "=&r" (tmp) \
| : "r" (v) \
| : cl); \
| \
| return i; \
| }
While others have negative indentation for some lines, and/or have
misaligned trailing backslashes, e.g.
| static inline void __lse_atomic_##op(int i, atomic_t *v) \
| { \
| asm volatile( \
| __LSE_PREAMBLE \
| " " #asm_op " %w[i], %[v]\n" \
| : [i] "+r" (i), [v] "+Q" (v->counter) \
| : "r" (v)); \
| }
This patch makes the indentation consistent and also aligns the trailing
backslashes. This makes the code easier to read for those (like myself)
who are easily distracted by these inconsistencies.
This is intended as a cleanup.
There should be no functional change as a result of this patch.
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Cc: Boqun Feng <boqun.feng@gmail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Will Deacon <will@kernel.org>
Acked-by: Will Deacon <will@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20211210151410.2782645-2-mark.rutland@arm.com
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
|
|
Ido Schimmel says:
====================
mlxsw: MAC profiles occupancy fix
Patch #1 fixes a router interface (RIF) MAC profiles occupancy bug that
was merged in the last cycle.
Patch #2 adds a selftest that fails without the fix.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Add a test case to cover the bug fixed by the previous patch.
Edit the MAC address of one netdev so that it matches the MAC address of
the second netdev. Verify that the two MAC profiles were consolidated by
testing that the MAC profiles occupancy decreased by one.
Signed-off-by: Danielle Ratson <danieller@nvidia.com>
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Currently, when setting a router interface (RIF) MAC address while the
MAC profile is not shared with other RIFs, the profile is edited so that
the new MAC address is assigned to it.
This does not take into account a situation in which the new MAC address
already matches an existing MAC profile. In that situation, two MAC
profiles will be occupied even though they hold MAC addresses from the
same profile.
In order to prevent that, add a check to ensure that editing a MAC
profile takes place only when the new MAC address does not match an
existing profile.
Fixes: 605d25cd782a6 ("mlxsw: spectrum_router: Add RIF MAC profiles support")
Reported-by: Maksym Yaremchuk <maksymy@nvidia.com>
Tested-by: Maksym Yaremchuk <maksymy@nvidia.com>
Signed-off-by: Danielle Ratson <danieller@nvidia.com>
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
__rds_conn_create() did not release conn->c_path when loop_trans != 0 and
trans->t_prefer_loopback != 0 and is_outgoing == 0.
Fixes: aced3ce57cd3 ("RDS tcp loopback connection can hang")
Signed-off-by: Hangyu Hua <hbh25y@gmail.com>
Reviewed-by: Sharath Srinivasan <sharath.srinivasan@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/jberg/mac80211
Johannes Berg says:
====================
A fairly large number of fixes this time:
* fix a station info memory leak on insert collisions
* a rate control fix for retransmissions
* two aggregation setup fixes
* reload current regdomain when reloading database
* a locking fix in regulatory work
* a probe request allocation size fix in mac80211
* apply TCP vs. aggregation (sk pacing) on mesh
* fix ordering of channel context update vs. station
state
* set up skb->dev for mesh forwarding properly
* track QoS data frames only for admission control to
avoid out-of-bounds read (found by syzbot)
* validate extended element ID vs. existing data to
avoid out-of-bounds read (found by syzbot)
* fix locking in mac80211 aggregation TX setup
* fix traffic stall after HW restart when TXQs are used
* fix ordering of reconfig/restart after HW restart
* fix interface type for extended aggregation capability
lookup
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/net-queue
Tony Nguyen says:
====================
Intel Wired LAN Driver Updates 2021-12-13
This series contains updates to iavf driver only.
Dan Carpenter fixes some missing mutex unlocking.
Stefan Assmann restores stopping watchdog from overriding to reset state.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
We need to return EOPNOTSUPP for the unsupported mpls action type when
setup the flow action.
In the original implement, we will return 0 for the unsupported mpls
action type, actually we do not setup it and the following actions
to the flow action entry.
Fixes: 9838b20a7fb2 ("net: sched: take rtnl lock in tc_setup_flow_action()")
Signed-off-by: Baowen Zheng <baowen.zheng@corigine.com>
Signed-off-by: Simon Horman <simon.horman@corigine.com>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
When CMD13 is sent after switching to HS400ES mode, the bus
is operating at either MMC_HIGH_26_MAX_DTR or MMC_HIGH_52_MAX_DTR.
To meet Tegra SDHCI requirement at HS400ES mode, force SDHCI
interface clock to MMC_HS200_MAX_DTR (200 MHz) so that host
controller CAR clock and the interface clock are rate matched.
Signed-off-by: Prathamesh Shete <pshete@nvidia.com>
Acked-by: Adrian Hunter <adrian.hunter@intel.com>
Fixes: dfc9700cef77 ("mmc: tegra: Implement HS400 enhanced strobe")
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20211214113653.4631-1-pshete@nvidia.com
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
|
|
To replicate the issue:-
1) Add 1 flower filter for VLAN Priority based frame steering:-
$ IFDEVNAME=eth0
$ tc qdisc add dev $IFDEVNAME ingress
$ tc qdisc add dev $IFDEVNAME root mqprio num_tc 8 \
map 0 1 2 3 4 5 6 7 0 0 0 0 0 0 0 0 \
queues 1@0 1@1 1@2 1@3 1@4 1@5 1@6 1@7 hw 0
$ tc filter add dev $IFDEVNAME parent ffff: protocol 802.1Q \
flower vlan_prio 0 hw_tc 0
2) Get the 'pref' id
$ tc filter show dev $IFDEVNAME ingress
3) Delete a specific tc flower record (say pref 49151)
$ tc filter del dev $IFDEVNAME parent ffff: pref 49151
From dmesg, we will observe kernel NULL pointer ooops
[ 197.170464] BUG: kernel NULL pointer dereference, address: 0000000000000000
[ 197.171367] #PF: supervisor read access in kernel mode
[ 197.171367] #PF: error_code(0x0000) - not-present page
[ 197.171367] PGD 0 P4D 0
[ 197.171367] Oops: 0000 [#1] PREEMPT SMP NOPTI
<snip>
[ 197.171367] RIP: 0010:tc_setup_cls+0x20b/0x4a0 [stmmac]
<snip>
[ 197.171367] Call Trace:
[ 197.171367] <TASK>
[ 197.171367] ? __stmmac_disable_all_queues+0xa8/0xe0 [stmmac]
[ 197.171367] stmmac_setup_tc_block_cb+0x70/0x110 [stmmac]
[ 197.171367] tc_setup_cb_destroy+0xb3/0x180
[ 197.171367] fl_hw_destroy_filter+0x94/0xc0 [cls_flower]
The above issue is due to previous incorrect implementation of
tc_del_vlan_flow(), shown below, that uses flow_cls_offload_flow_rule()
to get struct flow_rule *rule which is no longer valid for tc filter
delete operation.
struct flow_rule *rule = flow_cls_offload_flow_rule(cls);
struct flow_dissector *dissector = rule->match.dissector;
So, to ensure tc_del_vlan_flow() deletes the right VLAN cls record for
earlier configured RX queue (configured by hw_tc) in tc_add_vlan_flow(),
this patch introduces stmmac_rfs_entry as driver-side flow_cls_offload
record for 'RX frame steering' tc flower, currently used for VLAN
priority. The implementation has taken consideration for future extension
to include other type RX frame steering such as EtherType based.
v2:
- Clean up overly extensive backtrace and rewrite git message to better
explain the kernel NULL pointer issue.
Fixes: 0e039f5cf86c ("net: stmmac: add RX frame steering based on VLAN priority in tc flower")
Tested-by: Kurt Kanzenbach <kurt@linutronix.de>
Signed-off-by: Ong Boon Leong <boon.leong.ong@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
PCIe PMU Root Complex Integrated End Point(RCiEP) device is supported
to sample bandwidth, latency, buffer occupation etc.
Each PMU RCiEP device monitors multiple Root Ports, and each RCiEP is
registered as a PMU in /sys/bus/event_source/devices, so users can
select target PMU, and use filter to do further sets.
Filtering options contains:
event - select the event.
port - select target Root Ports. Information of Root Ports are
shown under sysfs.
bdf - select requester_id of target EP device.
trig_len - set trigger condition for starting event statistics.
trig_mode - set trigger mode. 0 means starting to statistic when bigger
than trigger condition, and 1 means smaller.
thr_len - set threshold for statistics.
thr_mode - set threshold mode. 0 means count when bigger than threshold,
and 1 means smaller.
Acked-by: Krzysztof Wilczyński <kw@linux.com>
Reviewed-by: John Garry <john.garry@huawei.com>
Signed-off-by: Qi Liu <liuqi115@huawei.com>
Reviewed-by: Shaokun Zhang <zhangshaokun@hisilicon.com>
Link: https://lore.kernel.org/r/20211202080633.2919-3-liuqi115@huawei.com
Signed-off-by: Will Deacon <will@kernel.org>
|
|
PCIe PMU Root Complex Integrated End Point(RCiEP) device is supported on
HiSilicon HIP09 platform. Document it to provide guidance on how to
use it.
Reviewed-by: John Garry <john.garry@huawei.com>
Signed-off-by: Qi Liu <liuqi115@huawei.com>
Reviewed-by: Shaokun Zhang <zhangshaokun@hisilicon.com>
Link: https://lore.kernel.org/r/20211202080633.2919-2-liuqi115@huawei.com
Signed-off-by: Will Deacon <will@kernel.org>
|
|
PCI_MSIX_FLAGS_MASKALL is set in the MSI-X control register at MSI-X
interrupt setup time. It's cleared on success, but the error handling path
only clears the PCI_MSIX_FLAGS_ENABLE bit.
That's incorrect as the reset state of the PCI_MSIX_FLAGS_MASKALL bit is
zero. That can be observed via lspci:
Capabilities: [b0] MSI-X: Enable- Count=67 Masked+
Clear the bit in the error path to restore the reset state.
Fixes: 438553958ba1 ("PCI/MSI: Enable and mask MSI-X early")
Reported-by: Stefan Roese <sr@denx.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Tested-by: Stefan Roese <sr@denx.de>
Cc: linux-pci@vger.kernel.org
Cc: Bjorn Helgaas <bhelgaas@google.com>
Cc: Michal Simek <michal.simek@xilinx.com>
Cc: Marek Vasut <marex@denx.de>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/87tufevoqx.ffs@tglx
|
|
Masking all unused MSI-X entries is done to ensure that a crash kernel
starts from a clean slate, which correponds to the reset state of the
device as defined in the PCI-E specificion 3.0 and later:
Vector Control for MSI-X Table Entries
--------------------------------------
"00: Mask bit: When this bit is set, the function is prohibited from
sending a message using this MSI-X Table entry.
...
This bit’s state after reset is 1 (entry is masked)."
A Marvell NVME device fails to deliver MSI interrupts after trying to
enable MSI-X interrupts due to that masking. It seems to take the MSI-X
mask bits into account even when MSI-X is disabled.
While not specification compliant, this can be cured by moving the masking
into the success path, so that the MSI-X table entries stay in device reset
state when the MSI-X setup fails.
[ tglx: Move it into the success path, add comment and amend changelog ]
Fixes: aa8092c1d1f1 ("PCI/MSI: Mask all unused MSI-X entries")
Signed-off-by: Stefan Roese <sr@denx.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-pci@vger.kernel.org
Cc: Bjorn Helgaas <bhelgaas@google.com>
Cc: Michal Simek <michal.simek@xilinx.com>
Cc: Marek Vasut <marex@denx.de>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20211210161025.3287927-1-sr@denx.de
|
|
Add device tree bindings for Last-level-cache Tag-and-data
(LLC-TAD) unit PMU for Marvell CN10K SoCs.
Signed-off-by: Bhaskara Budiredla <bbudiredla@marvell.com>
Reviewed-by: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/r/20211115043506.6679-3-bbudiredla@marvell.com
Signed-off-by: Will Deacon <will@kernel.org>
|
|
This driver adds support for Last-level cache tag-and-data unit
(LLC-TAD) PMU that is featured in some of the Marvell's CN10K
infrastructure silicons.
The LLC is divided into 2N slices distributed across N Mesh tiles
in a single-socket configuration. The driver always configures the
same counter for all of the TADs. The user would end up effectively
reserving one of eight counters in every TAD to look across all TADs.
The occurrences of events are aggregated and presented to the user
at the end of an application run. The driver does not provide a way
for the user to partition TADs so that different TADs are used for
different applications.
The event counters are zeroed to start event counting to avoid any
rollover issues. TAD perf counters are 64-bit, so it's not currently
possible to overflow event counters at current mesh and core
frequencies.
To measure tad pmu events use perf tool stat command. For instance:
perf stat -e tad_dat_msh_in_dss,tad_req_msh_out_any <application>
perf stat -e tad_alloc_any,tad_hit_any,tad_tag_rd <application>
Signed-off-by: Bhaskara Budiredla <bbudiredla@marvell.com>
Link: https://lore.kernel.org/r/20211115043506.6679-2-bbudiredla@marvell.com
Signed-off-by: Will Deacon <will@kernel.org>
|
|
Use the EOR3 instruction to implement xor_blocks() if the instruction is
available, which is the case if the CPU implements the SHA-3 extension.
This is about 20% faster on Apple M1 when using the 5-way version.
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
Link: https://lore.kernel.org/r/20211213140252.2856053-1-ardb@kernel.org
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
|
|
Livepatching a loaded module involves applying relocations through
apply_relocate_add(), which attempts to write to read-only memory when
CONFIG_STRICT_MODULE_RWX=y. Work around this by performing these
writes through the text poke area by using patch_instruction().
R_PPC_REL24 is the only relocation type generated by the kpatch-build
userspace tool or klp-convert kernel tree that I observed applying a
relocation to a post-init module.
A more comprehensive solution is planned, but using patch_instruction()
for R_PPC_REL24 on should serve as a sufficient fix.
This does have a performance impact, I observed ~15% overhead in
module_load() on POWER8 bare metal with checksum verification off.
Fixes: c35717c71e98 ("powerpc: Set ARCH_HAS_STRICT_MODULE_RWX")
Cc: stable@vger.kernel.org # v5.14+
Reported-by: Joe Lawrence <joe.lawrence@redhat.com>
Signed-off-by: Russell Currey <ruscur@russell.cc>
Tested-by: Joe Lawrence <joe.lawrence@redhat.com>
[mpe: Check return codes from patch_instruction()]
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20211214121248.777249-1-mpe@ellerman.id.au
|
|
The SMMU_PMCG_IIDR register was not present in older revisions of the
Arm SMMUv3 spec. On Arm Ltd. implementations, the IIDR value consists of
fields from several PIDR registers, allowing us to present a
standardized identifier to userspace.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Jean-Philippe Brucker <jean-philippe@linaro.org>
Link: https://lore.kernel.org/r/20211117144844.241072-4-jean-philippe@linaro.org
Signed-off-by: Will Deacon <will@kernel.org>
|
|
Add device-tree support to the SMMUv3 PMCG driver.
Signed-off-by: Jay Chen <jkchen@linux.alibaba.com>
Signed-off-by: Jean-Philippe Brucker <jean-philippe@linaro.org>
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Link: https://lore.kernel.org/r/20211117144844.241072-3-jean-philippe@linaro.org
Signed-off-by: Will Deacon <will@kernel.org>
|
|
Add binding for the Arm SMMUv3 PMU. Each node represents a PMCG, and is
placed as a sibling node of the SMMU. Although the PMCGs registers may
be within the SMMU MMIO region, they are separate devices, and there can
be multiple PMCG devices for each SMMU (for example one for the TCU and
one for each TBU).
Signed-off-by: Jean-Philippe Brucker <jean-philippe@linaro.org>
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Link: https://lore.kernel.org/r/20211117144844.241072-2-jean-philippe@linaro.org
Signed-off-by: Will Deacon <will@kernel.org>
|
|
In general, detailed performance analysis will require knoweldge of the
the SoC beyond the CMN itself - e.g. which actual CPUs/peripherals/etc.
are connected to each node. However for certain development and bringup
tasks it can be useful to have a quick overview of the CMN internal
topology to hand too. Add a debugfs file to map this out.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Link: https://lore.kernel.org/r/159fd4d7e19fb3c8801a8cb64ee73ec50f55903c.1638530442.git.robin.murphy@arm.com
Signed-off-by: Will Deacon <will@kernel.org>
|
|
Add the identifiers and events for the CI-700 coherent interconnect.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Link: https://lore.kernel.org/r/28f566ab23a83733c6c9ef9414c010b760b4549c.1638530442.git.robin.murphy@arm.com
Signed-off-by: Will Deacon <will@kernel.org>
|
|
CI-700 is a new client-level coherent interconnect derived from
the enterprise-level CMN family, and shares the same PMU design.
CC: devicetree@vger.kernel.org
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Reviewed-by: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/r/5f0b372f808f1468e6d9500cedafbecd10254674.1638530442.git.robin.murphy@arm.com
Signed-off-by: Will Deacon <will@kernel.org>
|
|
The second generation of CMN IPs add new node types and significantly
expand the configuration space with options for extra device ports on
edge XPs, either plumbed into the regular DTM or with extra dedicated
DTMs to monitor them, plus larger (and smaller) mesh sizes. Add basic
support for pulling this new information out of the hardware, piping
it around as necessary, and handling (most of) the new choices.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Link: https://lore.kernel.org/r/e58b495bcc7deec3882be4bac910ed0bf6979674.1638530442.git.robin.murphy@arm.com
Signed-off-by: Will Deacon <will@kernel.org>
|
|
In preparation for supporting newer CMN products, let's introduce a
means to differentiate the features and events which are specific to a
particular IP from those which remain common to the whole family. The
newer designs have also smoothed off some of the rough edges in terms
of discoverability, so separate out the parts of the flow which have
effectively now become CMN-600 quirks.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Link: https://lore.kernel.org/r/9f6368cdca4c821d801138939508a5bba54ccabb.1638530442.git.robin.murphy@arm.com
Signed-off-by: Will Deacon <will@kernel.org>
|
|
With the value of CMN_MAX_DTMS increasing significantly, our validation
data structure is set to get quite big. Technically we could pack it at
least twice as densely, since we only need around 19 bits of information
per DTM, but that makes the code even more mind-bogglingly impenetrable,
and even half of "quite big" may still be uncomfortably large for a
stack frame (~1KB). Just move it to an off-stack allocation instead.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Link: https://lore.kernel.org/r/0cabff2e5839ddc0979e757c55515966f65359e4.1638530442.git.robin.murphy@arm.com
Signed-off-by: Will Deacon <will@kernel.org>
|
|
In cases where we do know which DTC domain a node belongs to, we can
skip initialising or reading the global count in DTCs where we know
it won't change. The machinery to achieve that is mostly in place
already, so finish hooking it up by converting the vestigial domain
tracking to propagate suitable bitmaps all the way through to events.
Note that this does not allow allocating such an unused counter to a
different event on that DTC, because that is a flippin' nightmare.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Link: https://lore.kernel.org/r/51d930fd945ef51c81f5889ccca055c302b0a1d0.1638530442.git.robin.murphy@arm.com
Signed-off-by: Will Deacon <will@kernel.org>
|
|
When multiple nodes of the same type are connected to the same XP
(particularly in CAL configurations), it seems that they are likely
to be consecutive in logical ID. Therefore, we're likely to gain a
small benefit from an easy tweak to optimise out consecutive reads
of the same set of DTM counters for an aggregated event.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Link: https://lore.kernel.org/r/7777d77c2df17693cd3dabb6e268906e15238d82.1638530442.git.robin.murphy@arm.com
Signed-off-by: Will Deacon <will@kernel.org>
|
|
Untangle DTMs from XPs into a dedicated abstraction. This helps make
things a little more obvious and robust, but primarily paves the way
for further development where new IPs can grow extra DTMs per XP.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Link: https://lore.kernel.org/r/9cca18b1b98f482df7f1aaf3d3213e7f39500423.1638530442.git.robin.murphy@arm.com
Signed-off-by: Will Deacon <will@kernel.org>
|
|
Refactor the places where we scan through the set of nodes to switch
from explicit array indexing to pointer-based iteration. This leads to
slightly simpler object code, but also makes the source less dense and
more pleasant for further development. It also unearths an almost-bug
in arm_cmn_event_init() where we've been depending on the "array index"
of NULL relative to cmn->dns being a sufficiently large number, yuck.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Link: https://lore.kernel.org/r/ee0c9eda9a643f46001ac43aadf3f0b1fd5660dd.1638530442.git.robin.murphy@arm.com
Signed-off-by: Will Deacon <will@kernel.org>
|
|
Add a bit more abstraction for the places where we decompose node IDs.
This will help keep things nice and manageable when we come to add yet
more variables which affect the node ID format. Also use the opportunity
to move the rest of the low-level node management helpers back up to the
logical place they were meant to be - how they ended up buried right in
the middle of the event-related definitions is somewhat of a mystery...
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Link: https://lore.kernel.org/r/a2242a8c3c96056c13a04ae87bf2047e5e64d2d9.1638530442.git.robin.murphy@arm.com
Signed-off-by: Will Deacon <will@kernel.org>
|
|
Although CMN is currently (and overwhelmingly likely to remain) deployed
in arm64-only (modulo userspace) systems, the 64-bit "dependency" for
compile-testing was just laziness due to heavy reliance on readq/writeq
accessors. Since we only need one extra include for robustness in that
regard, let's pull that in, widen the compile-test coverage, and fix up
the smattering of type laziness that that brings to light.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Link: https://lore.kernel.org/r/baee9ee0d0bdad8aaeb70f5a4b98d8fd4b1f5786.1638530442.git.robin.murphy@arm.com
Signed-off-by: Will Deacon <will@kernel.org>
|
|
On a system with multiple CMN meshes, ideally we'd want to access each
PMU from within its own mesh, rather than with a long CML round-trip,
wherever feasible. Since such a system is likely to be presented as
multiple NUMA nodes, let's also hope a proximity domain is specified
for each CMN programming interface, and use that to guide our choice
of IRQ affinity to favour a node-local CPU where possible.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Link: https://lore.kernel.org/r/32438b0d016e0649d882d47d30ac2000484287b9.1638530442.git.robin.murphy@arm.com
Signed-off-by: Will Deacon <will@kernel.org>
|
|
Attempting to migrate the PMU context after we've unregistered the PMU
device, or especially if we never successfully registered it in the
first place, is a woefully bad idea. It's also fundamentally pointless
anyway. Make sure to unregister an instance from the hotplug handler
*without* invoking the teardown callback.
Fixes: 0ba64770a2f2 ("perf: Add Arm CMN-600 PMU driver")
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Link: https://lore.kernel.org/r/2c221d745544774e4b07583b65b5d4d94f7e0fe4.1638530442.git.robin.murphy@arm.com
Signed-off-by: Will Deacon <will@kernel.org>
|
|
Add documentation to describe the access to the pmu hardware counters from
userspace.
Signed-off-by: Raphael Gault <raphael.gault@arm.com>
Signed-off-by: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/r/20211208201124.310740-6-robh@kernel.org
Signed-off-by: Will Deacon <will@kernel.org>
|
|
Arm PMUs can support direct userspace access of counters which allows for
low overhead (i.e. no syscall) self-monitoring of tasks. The same feature
exists on x86 called 'rdpmc'. Unlike x86, userspace access will only be
enabled for thread bound events. This could be extended if needed, but
simplifies the implementation and reduces the chances for any
information leaks (which the x86 implementation suffers from).
PMU EL0 access will be enabled when an event with userspace access is
part of the thread's context. This includes when the event is not
scheduled on the PMU. There's some additional overhead clearing
dirty counters when access is enabled in order to prevent leaking
disabled counter data from other tasks.
Unlike x86, enabling of userspace access must be requested with a new
attr bit: config1:1. If the user requests userspace access with 64-bit
counters, then the event open will fail if the h/w doesn't support
64-bit counters. Chaining is not supported with userspace access. The
modes for config1 are as follows:
config1 = 0 : user access disabled and always 32-bit
config1 = 1 : user access disabled and always 64-bit (using chaining if needed)
config1 = 2 : user access enabled and always 32-bit
config1 = 3 : user access enabled and always 64-bit
Based on work by Raphael Gault <raphael.gault@arm.com>, but has been
completely re-written.
Cc: Will Deacon <will@kernel.org>
Cc: Mark Rutland <mark.rutland@arm.com>
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: Catalin Marinas <catalin.marinas@arm.com>
Cc: linux-arm-kernel@lists.infradead.org
Cc: linux-perf-users@vger.kernel.org
Signed-off-by: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/r/20211208201124.310740-5-robh@kernel.org
[will: Made armv8pmu_proc_user_access_handler() static]
Signed-off-by: Will Deacon <will@kernel.org>
|
|
Like x86, some users may want to disable userspace PMU counter
altogether. Add a sysctl 'perf_user_access' file to control userspace
counter access. The default is '0' which is disabled. Writing '1'
enables access.
Note that x86 supports globally enabling user access by writing '2' to
/sys/bus/event_source/devices/cpu/rdpmc. As there's not existing
userspace support to worry about, this shouldn't be necessary for Arm.
It could be added later if the need arises.
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: Catalin Marinas <catalin.marinas@arm.com>
Cc: linux-perf-users@vger.kernel.org
Acked-by: Will Deacon <will@kernel.org>
Reviewed-by: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/r/20211208201124.310740-4-robh@kernel.org
Signed-off-by: Will Deacon <will@kernel.org>
|
|
On arm64, user space counter access will be controlled differently
compared to x86. On x86, access in the strictest mode is enabled for all
tasks in an MM when any event is mmap'ed. For arm64, access is
explicitly requested for an event and only enabled when the event's
context is active. This avoids hooks into the arch context switch code
and gives better control of when access is enabled.
In order to configure user space access when the PMU is enabled, it is
necessary to know if any event (currently active or not) in the current
context has user space accessed enabled. Add a counter similar to other
counters in the context to avoid walking the event list every time.
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/20211208201124.310740-3-robh@kernel.org
Signed-off-by: Will Deacon <will@kernel.org>
|
|
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>
|
