An IPI sent to the local hart can be processed directly instead of
triggering the IPI device. This is more efficient, and it avoids a
deadlock when the .sync callback is defined. Since interrupts are
disabled while handling an ecall, the IPI would not get delivered
until the next mret, but sbi_ipi_sync() is called before then.
Signed-off-by: Samuel Holland <samuel.holland@sifive.com>
Reviewed-by: Anup Patel <anup@brainfault.org>
Reviewed-by: Xiang W <wxjstz@126.com>
Currently, failures in sbi_ipi_send() are silently ignored, which makes
them difficult to debug. Instead, abort sending the IPI and pass back
the error, but still synchronize any IPIs already sent.
Signed-off-by: Samuel Holland <samuel.holland@sifive.com>
Signed-off-by: Anup Patel <anup@brainfault.org>
The tlb_update() callback is called for each destination hart.
Move the size check earlier, so it is executed only once.
Signed-off-by: Samuel Holland <samuel.holland@sifive.com>
Reviewed-by: Anup Patel <anup@brainfault.org>
Currently, the fdt_parse_aclint_node() follows a fixed order to parse
ACLINT timer. This may cause the undesirable result when the ACLINT
device does not support mtime without adding an empty entry for it in
the DT.
To be robust, make fdt_parse_aclint_node() support "reg-names" property,
so it can parse the DT in an order independent way. For compatibility,
fdt_parse_aclint_node() only use "reg-names" when parsing ACLINT timer,
and will fallback to the old way if "reg-names" property is not found.
Link: https://lore.kernel.org/all/20231114-skedaddle-precinct-66c8897227bb@squawk/
Signed-off-by: Inochi Amaoto <inochiama@outlook.com>
Reviewed-by: Anup patel <anup@brainfault.org>
When warmboot via HSM, coolboot has been completed and
wait_for_coldboot can be skipped to speed up.
Signed-off-by: Xiang W <wxjstz@126.com>
Reviewed-by: Anup Patel <anup@brainfault.org>
Previous patch introduced a change that using hart count as the default
number of tlb entries in the fifo. This makes the default tlb fifo size
grow in square with the number of harts. So the default heap size is
not enough to allocate tlb fifo when the hart count is big.
Fixes: 52fd64b ("platform: Uses hart count as the default size of tlb info")
Signed-off-by: Inochi Amaoto <inochiama@outlook.com>
Reviewed-by: Anup Patel <anup@brainfault.org>
Current interruptible hart mask doesn't include the hart which HSM state
is SBI_HSM_STATE_RESUME_PENDING. So when there is a request to send an
IPI to the hart which is in the resume process, this hart would miss the
IPI forever. Put the SBI_HSM_STATE_RESUME_PENDING hart in the
interruptible hart mask to fix the issue.
Signed-off-by: Nick Hu <nick.hu@sifive.com>
Reviewed-by: Xiang W <wxjstz@126.com>
Reviewed-by: Anup Patel <anup@brainfault.org>
The counter info should only return valid hardware counters for the ones
set in the counter mask. Otherwise, it will report incorrect number of
hardware counters to the supervisor if the platform has discontiguous
counters.
Fixes: c744ed77b1 ("lib: sbi_pmu: Enable noncontigous hpm event and counters")
Signed-off-by: Atish Patra <atishp@rivosinc.com>
Reviewed-by: Xiang W <wxjstz@126.com>
Reviewed-by: Anup Patel <anup@brainfault.org>
OpenSBI doesn't support SBI PMU snapshot yet as there is not much benefit
unless the multiple counters overflow at the same time.
Just add the definition and return not supported error at this moment. The
default returned error is also not supported. Thus, no functional change
intended.
Reviewed-by: Anup Patel <anup@brainfault.org>
Signed-off-by: Atish Patra <atishp@rivosinc.com>
Multiple harts may try to send IPI to a particular target hart A
in which case the send_ipi() should be called only when the old
value of the hart A ipi_type is zero.
Signed-off-by: Xiang W <wxjstz@126.com>
Reviewed-by: Anup Patel <anup@brainfault.org>
Simplify atomic-related bit operations through __atomic intrinsics.
Signed-off-by: Xiang W <wxjstz@126.com>
Reviewed-by: Anup Patel <anup@brainfault.org>
The original code returns the value of the word before modification.
When modifying the upper 32 bits under RV64, the value returned via
int return will have no meaning. Corrected to return the value of the
bit. And modify the function description.
Signed-off-by: Xiang W <wxjstz@126.com>
Reviewed-by: Anup Patel <anup@brainfault.org>
lr/sc is part of the A extension. If the A extension is not supported,
lr/sc cannot be used. So remove xchg/cmpxchg.
Signed-off-by: Xiang W <wxjstz@126.com>
Reviewed-by: Anup Patel <anup@brainfault.org>
Add PMU node example for event index to counter index mapping
and selector value translation of Andes' CPUs.
Currently, there are 4 HPM counters that can be used to monitor
all of the events for each hart.
Signed-off-by: Yu Chien Peter Lin <peterlin@andestech.com>
Reviewed-by: Locus Wei-Han Chen <locus84@andestech.com>
Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
Reviewed-by: Anup Patel <anup@brainfault.org>
Reviewed-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
The valid entry count is tracking by hw_event_count so there
is no need to check the whole table.
Signed-off-by: Yu Chien Peter Lin <peterlin@andestech.com>
Reviewed-by: Atish Patra <atishp@rivosinc.com>
Reviewed-by: Anup Patel <anup@brainfault.org>
To allow platform override pmu_init() filling the translation table
fdt_pmu_evt_select[] when PMU node doesn't provide such information,
we need to share the table and its entry counter with other .c file.
We also define the structures of PMU property in fdt_helper.h, so we
can initialize the mappings in arrays.
Signed-off-by: Yu Chien Peter Lin <peterlin@andestech.com>
Reviewed-by: Atish Patra <atishp@rivosinc.com>
Reviewed-by: Anup Patel <anup@brainfault.org>
We will need is_andes(45) in the following patch,
so factor out the code that parses marchid to make
it reusable for checking any Andes CPU variants.
Also improves the comment in ae350_hart_start().
Signed-off-by: Yu Chien Peter Lin <peterlin@andestech.com>
Reviewed-by: Anup Patel <anup@brainfault.org>
Add a Kconfig option to control PMU fixup, so the next
stage software can dump the PMU node including event
mapping information for debugging purposes.
Signed-off-by: Yu Chien Peter Lin <peterlin@andestech.com>
Reviewed-by: Atish Patra <atishp@rivosinc.com>
Reviewed-by: Anup Patel <anup@brainfault.org>
Before the ratification of Sscofpmf, the Andes PMU extension
was designed to support the sampling and filtering with hardware
performance counters (zihpm), it works with the current SBI PMU
extension and Linux SBI PMU driver.
We implement 1) the PMU device callbacks that update the
corresponding bits on custom CSRs, 2) extentions_init() to detect
the hardware support of Andes PMU and initialize the per-hart
PMU related CSR, and 3) pmu_init() to register PMU device and
populate event mappings.
Also define a andes_pmu_setup() function which is in preparation
for adding default PMU mappings in andes_hpm.h
Signed-off-by: Yu Chien Peter Lin <peterlin@andestech.com>
Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
Reviewed-by: Anup Patel <anup@brainfault.org>
Add support for custom PMU extensions to set inhibit bits
on custom CSRs by introducing the PMU device callback
hw_counter_filter_mode(). This allows the perf tool to
restrict event counting under a specified privileged
mode by appending a modifier, e.g. perf record -e event:k
to count events only happening in kernel mode.
Signed-off-by: Yu Chien Peter Lin <peterlin@andestech.com>
Reviewed-by: Leo Yu-Chi Liang <ycliang@andestech.com>
Reviewed-by: Atish Patra <atishp@rivosinc.com>
Reviewed-by: Anup Patel <anup@brainfault.org>
This patch makes the following changes:
- As sbi_platform_pmu_init() returns a negative error code on
failure, let sbi_pmu_init() print out the error code with
sbi_dprintf().
- In order to distinguish the SBI_EFAIL error returned by
sbi_pmu_add_*_counter_map(), return SBI_ENOENT to indicate
that fdt_pmu_setup() failed to locate "riscv,pmu" node, and
generic_pmu_init() ignores such case.
Signed-off-by: Yu Chien Peter Lin <peterlin@andestech.com>
Reviewed-by: Anup Patel <anup@brainfault.org>
Reviewed-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
Reviewed-by: Atish Patra <atishp@rivosinc.com>
The old scheme doesn't allow sending hart0 self-IPI as the
corresponding bit on pending register is hardwired to 0, this
could lead to unhandle IPIs on SMP systems, esp. on single-core.
Furthermore, the limitation of old scheme is 8-core, instead of
reserving source hart information, we assign bit (x + 1) as the
enable and pending bit of hartx, this also expands the bootable
hart number.
The following diagram shows the enable bits of the new scheme
on 32-core Andes platform.
Pending regs: 0x1000 x---0---0---0---0------0---0
Pending hart ID: 0 1 2 3 ... 30 31
Interrupt ID: 0 1 2 3 4 ... 31 32
| | | | | | |
Enable regs: 0x2000 x---1---0---0---0-...--0---0---> hart0
| | | | | | |
0x2080 x---0---1---0---0-...--0---0---> hart1
| | | | | | |
0x2100 x---0---0---1---0-...--0---0---> hart2
| | | | | | |
0x2180 x---0---0---0---1-...--0---0---> hart3
. . . . . . .
. . . . . . .
. . . . . . .
0x2f00 x---0---0---0---0-...--1---0---> hart30
| | | | | | |
0x2f80 x---0---0---0---0-...--0---1---> hart31
<-------- word 0 -------><--- word 1 --->
To send IPI to hart0, for example, another hart (including hart0
itself) will set bit 1 of first word on the pending register.
We also fix indentation in andes_plicsw.h along with this patch.
Fixes: ce7c490719 ("lib: utils/ipi: Add Andes fdt ipi driver support")
Signed-off-by: Leo Yu-Chi Liang <ycliang@andestech.com>
Reviewed-by: Yu Chien Peter Lin <peterlin@andestech.com>
Reviewed-by: Randolph <randolph@andestech.com>
Reported-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
Link: https://lists.infradead.org/pipermail/opensbi/2023-October/005665.html
Reviewed-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
Tested-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
Reviewed-by: Anup Patel <anup@brainfault.org>
Currently, there are no barriers before or after the ipi_clear()
device callback which forces ipi_clear() device callback to always
use non-relaxed MMIO writes.
Instead of above, we use wmb() in after the ipi_clear() device
callback which pairs with the wmb() done before the ipi_send()
device callback. This also allows device ipi_clear() callback
to use relaxed MMIO writes.
Signed-off-by: Anup Patel <apatel@ventanamicro.com>
Reported-by: Bo Gan <ganboing@gmail.com>
Currently, we have a smp_wmb() between atomic_raw_set_bit() and
ipi_send() device callback whereas the MMIO writes done by the
device ipi_send() callback will also include a barrier.
We can avoid unnecessary/redundant barriers described above by
allowing relaxed MMIO writes in device ipi_send() callback. To
achieve this, we simply use wmb() instead of smp_wmb() before
calling device ipi_send().
Signed-off-by: Anup Patel <apatel@ventanamicro.com>
Reported-by: Bo Gan <ganboing@gmail.com>
Each APLIC CLRIE register allows disabling 32 interrupt sources at
a time by writing -1 so no need to write CLRIE register separately
for each interrupt source.
Fixes: 99792653de ("lib: utils/irqchip: Add APLIC initialization library")
Signed-off-by: Anup Patel <apatel@ventanamicro.com>
Reviewed-by: Atish Patra <atishp@rivosinc.com>
total_ctrs is bounded by
SBI_PMU_FW_CTR_MAX + SBI_PMU_HW_CTR_MAX) == 48
which exceeds BITS_PER_LONG on 32 bit systems.
Iterating over the bits of &cmask results in a buffer overflow when looking
for a bit >= BITS_PER_LONG.
Adjust the iterators in sbi_pmu_ctr_start() and sbi_pmu_ctr_stop()
accordingly.
Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
Reviewed-by: Anup Patel <anup@brainfault.org>
The sbi_hart_map_saddr() must create PMP mapping of size greater
than or equal to PMP granularity otherwise PMP mapping does not
work when size parameter less than sbi_hart_pmp_granularity(scratch).
Fixes: 6e44ef686a ("lib: sbi: Add functions to map/unmap shared memory")
Signed-off-by: Anup Patel <apatel@ventanamicro.com>
Reviewed-by: Samuel Holland <samuel.holland@sifive.com>
For some debuggers that do not implement SYSWRITEC and SYSREADC
operations, we have to use SYSWRITE and SYSREAD.
Instead of implementing semihosting_putc() using SYSWRITE, let us
simply remove semihosting_putc() because console_putc/console_puts
are now interchangeable.
Signed-off-by: Chen Pei <cp0613@linux.alibaba.com>
Signed-off-by: Xiang W <wxjstz@126.com>
Reviewed-by: Guo Ren <guoren@kernel.org>
Reviewed-by: Anup Patel <anup@brainfault.org>
console_puts/console_putc should replace each other, but the previous
sbi_putc can only use console_putc. This patch addresses this problem.
Signed-off-by: Xiang W <wxjstz@126.com>
Reviewed-by: Guo Ren <guoren@kernel.org>
Reviewed-by: Anup Patel <anup@brainfault.org>
In the future there may be a lot of ISA extensions, a 'long' may not
be able to accommodate, changed to an array for the future.
Addresses-Coverity-ID: 1568357 Out-of-bounds access
Fixes: 6259b2ec2d ("lib: utils/fdt: Fix fdt_parse_isa_extensions()
implementation")
Signed-off-by: Xiang W <wxjstz@126.com>
Reviewed-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
Reviewed-by: Anup Patel <anup@brainfault.org>
If ISA extension Zkr is available, set
mseccfg.sseed=1
mseccfg.useed=0
This enables access to the seed CSR in S-mode but not in U-mode.
Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
Reviewed-by: Anup Patel <anup@brainfault.org>
- Add Zkr as extension in sbi_hart_extensions enum
- Return "zkr" string for Zkr extension from sbi_hart_extension_id2string
Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
Reviewed-by: Anup Patel <anup@brainfault.org>
Define macros to access the sseed and the useed bit in the machine
security configuration register (mseccfg).
Signed-off-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
Reviewed-by: Xiang W <wxjstz@126.com>
Reviewed-by: Anup Patel <anup@brainfault.org>
As the domain will reject a new memory region which has a sub-regions
already in the domain, even the new region is bigger and has the same
flags. This problem can be solved by relaxing region restriction and
rechecking when adding and sanitizing domains.
Signed-off-by: Inochi Amaoto <inochiama@outlook.com>
Reviewed-by: Anup Patel <anup@brainfault.org>
Swapping domain region is a common operation when sorting domain region,
so separate it as a function to make code clean.
Signed-off-by: Inochi Amaoto <inochiama@outlook.com>
Reviewed-by: Anup Patel <anup@brainfault.org>
The quirks checking will cause ACLINT step into a CLINT code path, this
is not expected when ACLINT needs custom quirks.
Add a new quirk to identify custom ACLINT, and apply the general quirks
after applying CLINT specific quirks.
Signed-off-by: Inochi Amaoto <inochiama@outlook.com>
Reviewed-by: Anup Patel <anup@brainfault.org>
In the past, we used fdt_reset_thead to help customers with prototype
verification. However, with the emergence of the Big-little SoC system,
it can no longer meet the demand. Therefore, we use zero_stage_boot
instead of fdt_reset_thead. It cleans up the opensbi code and ends the
disputation of reset_sample's dts.
This patch removes the fdt_reset_thead component and updates the related
doc.
Signed-off-by: Guo Ren <guoren@linux.alibaba.com>
Signed-off-by: Guo Ren <guoren@kernel.org>
Reviewed-by: Anup Patel <anup@brainfault.org>
It's wrong to put the fence after setting the boot status flag because
all relocation operations must be finished before setting the status
flag. So, this fence must be put before the setting status flag, and
there is no use in putting a fence between _start_warm and setting
status flag.
Also, nop can't delay other harts too much, so use div instead, just
like Linux cpu_relax. Current opensbi force enables “M” Standard
Extension, and mul instructions have been used in the fw_base.S.
After the above two fixes, the boot hart index param of the
fw_dynamic_info could be guaranteed properly for all platforms.
Signed-off-by: Guo Ren <guoren@linux.alibaba.com>
Signed-off-by: Guo Ren <guoren@kernel.org>
Reviewed-by: Anup Patel <anup@brainfault.org>
For some debuggers that do not implement SYSWRITEC and SYSREADC
operations, we can use SYSWRITE and SYSREAD instead like the
implementation of semihosting_getc().
Signed-off-by: Chen Pei <cp0613@linux.alibaba.com>
Reviewed-by: Xiang W <wxjstz@126.com>
Reviewed-by: Anup Patel <anup@brainfault.org>
For platform with high number of harts, it is better to auto detect a
suitable number of entries in tlb fifo. Since allocating tlb entry for
all online harts can reduce the wait time significantly, using the
number of the online harts can make most platforms happy. This auto
detection can avoid most duplicate code for setting tlb fifo size.
Signed-off-by: Inochi Amaoto <inochiama@outlook.com>
Acked-by: Guo Ren <guoren@kernel.org>
Reviewed-by: Xiang W <wxjstz@126.com>
Reviewed-by: Anup Patel <anup@brainfault.org>
