This is needed when building HWPM driver locally without
NVIDIA build system.
- driver.c: use kernel version check to select the correct
signature of class:devnode
- mem_mgmt_utils.c:
- use kernel version check to provide correct parameter
of MODULE_IMPORT_NS macro
- use kernel version check to select the correct signature
of get_user_pages function
- mem_mgmt_utils.h: use kernel version check to select
between iosys-map or dma-buf-map
Bug 5120925
Change-Id: Ib33afc4d99056d5b872f0d4362e0e6c25eb7b64a
Signed-off-by: Besar Wicaksono <bwicaksono@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-hwpm/+/3306471
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: Yifei Wan <ywan@nvidia.com>
Reviewed-by: Vasuki Shankar <vasukis@nvidia.com>
The HWPM driver fails to be with Linux v6.13 because of the following
two issues:
1. In Linux v6.13, commit cdd30ebb1b9f ("module: Convert symbol
namespace to string literal") updated the MODULE_IMPORT_NS macro to
take a string literal as an argument in Linux v6.13. Use conftest to
detect if MODULE_IMPORT_NS takes a string literal as an argument and
update the HWPM driver accordingly.
2. The following build error is observed:
In file included from os/linux/clk_rst_utils.c:17:
include/linux/reset.h:30:49:
error: implicit declaration of function ‘BIT’
[-Werror=implicit-function-declaration]
30 | #define RESET_CONTROL_FLAGS_BIT_ACQUIRED BIT(2)
| ^~~
Fix the above by including the 'bits.h' header file.
Bug 4991705
Change-Id: I26cba920a0b0af251fd2f623ab9326ecafef5a5f
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-hwpm/+/3261738
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: Besar Wicaksono <bwicaksono@nvidia.com>
Reviewed-by: Vasuki Shankar <vasukis@nvidia.com>
Currently, HWPM driver checks regops address to validate that the
address belongs to an IP allowlist that is reserved for profiling.
However, it is possible that the allowlist doesn't include all register
offset that are required for profiling. This scenario is often
encountered during early stages of bringup. This patch adds a debugfs
node to make HWPM driver skip allowlist check. This change will allow
users to dynamically skip allowlist check when debugfs is available.
JIRA THWPM-65
Change-Id: Ic85a1c7fac6a95f7cde532f3bdf6040bbcc7f5f3
Signed-off-by: Vedashree Vidwans <vvidwans@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-hwpm/+/3241080
Reviewed-by: Seema Khowala <seemaj@nvidia.com>
Reviewed-by: Vasuki Shankar <vasukis@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: Vasuki Shankar <vasukis@nvidia.com>
The num_entries variable indicates the number of credit
programming requests that is sent from user space test app.
Current implementation has been configured to loop through
this in such a way that leads to an additional loop.
This change fixes the issue.
Bug 4571175
Signed-off-by: vasukis <vasukis@nvidia.com>
Change-Id: Id9fd4315e5ef470c697bf2815e16b42e746edf45
Reviewed-on: https://git-master.nvidia.com/r/c/linux-hwpm/+/3212369
Reviewed-by: Seema Khowala <seemaj@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Add IOCTL infra for Cross trigger programming in HWPM Driver.
Cross Triggering involves the access to secure register, which
cannot be issued by user space application. Hence, implement
cross trigger functionality in HWPM kernel driver.
Bug 4571175
Signed-off-by: vasukis <vasukis@nvidia.com>
Change-Id: Ia46227c4678d3ee282ebae8c58e116feaf4e59cb
Reviewed-on: https://git-master.nvidia.com/r/c/linux-hwpm/+/3147289
Reviewed-by: Seema Khowala <seemaj@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
HWPM driver uses nested structures and arrays of structures. The IP
structure setup logic allocates pointer arrays based on dynamic list of
IPs and aperture addresses. This dynamic list is required to search
given regops address in less amount of time.
However, there is a chance that the number of pointers computed
dynamically is huge. And huge amount of memory will be required for the
dynamic pointers array, which is impractical.
This, this patch modifies ip structure setup and address to aperture
conversion logic to use static indexes if the pointer array size is
huge.
This patch modifies relevant functions to always use static arrays
to access instance and aperture structures.
If dynamic pointers array is allocated, the patch adds logic to
translate dynamic index to static index using inst_index_mask for
instances and new added aperture_index for element level structures.
Add/update few log message to improve relayed information.
Bug 4707244
Change-Id: Ib4847e6575f82b628a3ce838ad69196a4bc08fed
Signed-off-by: Vedashree Vidwans <vvidwans@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-hwpm/+/3186843
Reviewed-by: Vasuki Shankar <vasukis@nvidia.com>
Reviewed-by: Seema Khowala <seemaj@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
This patch fixes issues found during testing
and guidance provided by devtools. The following
is changed in this patch:
1. mcf_iobhx and mcf_ocu are merged into a single mcf_soc IP.
2a. c2c is changed from 2 instances to 1.
2b. Remove C2CS0/1 which are the broadcast apertures.
Also remove the allowlist offset specific to broadcast
aperture.
3. mss_hub is changed from 1 instance to 8.
4. mss_channel is changed from 1 instance to 32.
5. mc0 perfmux is added to mcf_clink.
6. mcf_core is changed from 1 instance to 8.
7. License headers updated where necessary.
8. c2c allowlist updated to have just the offsets common
to all links.
9. Added a verbose comment explaining the design of
th500_hwpm_force_enable_ips()
10. Added back validate_current_config module parameter
as many systems still don't support fuses.
11. If all F's are read back for a regop in ip_readl(),
return -ENODEV.
There is a corresponding patch to update the python scripts
that generated many of the C and header files.
Bug 4287384
Change-Id: I8e14b0165dfa1abb9f5e04de577a41f0eb278246
Signed-off-by: Vishal Aslot <vaslot@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-hwpm/+/3134365
Reviewed-by: Seema Khowala <seemaj@nvidia.com>
Reviewed-by: Eric Lu <ericlu@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
ACPI configs are not required for aaos as
aaos is booted with dtb and not acpi.
Disabling CONFIG_ACPI results in a build failure as
it's undefined in the #if directive
so adding this check to see whether it is defined
resolves the issue and allows aaos to build
successfully with CONFIG_ACPI disabled
Bug 4559177
Change-Id: I9f068c373d6dc57acb610a107eb8a2e90a0e944b
Signed-off-by: Ahmad Chaudhry <ahmadc@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-hwpm/+/3115456
Reviewed-by: Vasuki Shankar <vasukis@nvidia.com>
GVS: Gerrit_Virtual_Submit <buildbot_gerritrpt@nvidia.com>
Reviewed-by: Vedashree Vidwans <vvidwans@nvidia.com>
HWPM code from HWPM repo is not currently used to compile with kernel
5.10. However, CL to compile HWPM repo for kernel 5.10 is required to
validate latest changes on Pre-Si.
Since conftest is only available for kernel version later than 5.10.
Add condition to include conftest only if HWPM is used as OOT module.
Bug 4119327
Change-Id: I760164447ff5c340884212f83966af72f1ee27da
Signed-off-by: Vedashree Vidwans <vvidwans@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-hwpm/+/3011333
Tested-by: Vasuki Shankar <vasukis@nvidia.com>
Reviewed-by: Vasuki Shankar <vasukis@nvidia.com>
Reviewed-by: Jonathan Hunter <jonathanh@nvidia.com>
Reviewed-by: Vishal Aslot <vaslot@nvidia.com>
Reviewed-by: Seema Khowala <seemaj@nvidia.com>
GVS: Gerrit_Virtual_Submit <buildbot_gerritrpt@nvidia.com>
The conftest script already has a test for checking which variant of the
get_user_pages() function is present in the kernel. So use the
definition generated by conftest to select which function variant is
used.
Bug 4276500
Change-Id: I29d216c8cead657c1daca4ce11b3dc3f74928467
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-hwpm/+/3015357
Reviewed-by: Vedashree Vidwans <vvidwans@nvidia.com>
GVS: Gerrit_Virtual_Submit <buildbot_gerritrpt@nvidia.com>
Currently, HWPM requires raw readl/writel functions
to access sticky bits and as workaround for IP registers.
- Move the raw readl/writel logic along with IO mapping
of the address to a static function.
- Implement the wrapper functions to access sticky bits
and IP registers to use the created static functions.
Jira THWPM-86
Change-Id: Ib0b3229d4b8795d19aca142233622a166436e3bd
Signed-off-by: Vedashree Vidwans <vvidwans@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-hwpm/+/3014028
Reviewed-by: Adeel Raza <araza@nvidia.com>
Reviewed-by: Vasuki Shankar <vasukis@nvidia.com>
Reviewed-by: Seema Khowala <seemaj@nvidia.com>
GVS: Gerrit_Virtual_Submit <buildbot_gerritrpt@nvidia.com>
This patch adds support for MCF SOC performance
monitoring in the driver. MCF SOC has two different
types of perfmuxes connected to the same perfmon:
one is the OCU type and the other is IBHX and OBHX.
IBHX is only accessible via MC16 aperture. Therefore,
this patch adds two separate IPs: OCU and IOBHX.
However, both are tied to the MCF SOC perfmon (mcfsoc0).
Bug 4287384
Signed-off-by: Vishal Aslot <vaslot@nvidia.com>
Change-Id: If15498a44e02270f9106337078931edbe043c254
Reviewed-on: https://git-master.nvidia.com/r/c/linux-hwpm/+/2986232
GVS: Gerrit_Virtual_Submit <buildbot_gerritrpt@nvidia.com>
Reviewed-by: Vedashree Vidwans <vvidwans@nvidia.com>
This patch adds support for C-NVLINK performance
monitoring in the driver. C-NVLINK consists of
RX, TX, and CTRL apertures, each with its own
perfmux signals and perfmons. So this patch
breaks them up into three sets of perfmux-perfmon
data structures.
Bug 4287384
Signed-off-by: Vishal Aslot <vaslot@nvidia.com>
Change-Id: Id8be4c965018125765f75a7b8bc8ab809bb7f976
Reviewed-on: https://git-master.nvidia.com/r/c/linux-hwpm/+/2999166
GVS: Gerrit_Virtual_Submit <buildbot_gerritrpt@nvidia.com>
Reviewed-by: Vedashree Vidwans <vvidwans@nvidia.com>
In Linux v6.2, the 'struct class.devnode()' function was updated to take
a 'const struct device *' instead of a 'struct device *'. A test has
been added to the conftest script to check for this and so instead of
relying on kernel version, use the definition generated by conftest to
select the appropriate function to use.
This is beneficial for working with 3rd party Linux kernels that may
have back-ported upstream changes into their kernel and so the kernel
version checks do not work.
Bug 4119327
Change-Id: I751b7401adee7b337192e255253b974cbd803642
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-hwpm/+/2991966
(cherry picked from commit 4b2fd8250d)
Reviewed-on: https://git-master.nvidia.com/r/c/linux-hwpm/+/2995574
GVS: Gerrit_Virtual_Submit <buildbot_gerritrpt@nvidia.com>
- HALs get_rtr_int_idx and get_ip_max_idx return the chip specific
router index and number of IPs. This information is static for a chip
and doesn't require any input. Hence, update the HAL definition to not
require hwpm pointer as an argument. Update definition and references
for these HALs.
- Add new HAL to get PMA and RTR structure pointers. Implement and
update other chip specific functions to use new HAL.
- Add new timer macro to check a condition and timeout after given
retries. Update necessary code to use new timer macro.
- Correct validate_emc_config function to compute correct available mss
channel mask based on fuse value.
- Update tegra_hwpm_readl and tegra_hwpm_writel macros to assert error
value. This way error checks are added at one spot and not sprinkled all
over the driver code.
- Update get_mem_bytes_put_ptr() and membuf_overflow_status() to return
error as function return and accept arguments to return mem_head pointer
and overflow status respectively. Add overflow status macros to use
throughout driver. Update HAL definition and references accordingly.
- conftest is only compiled for OOT config atm. Add OOT config check to
include conftest header.
Jira THWPM-109
Change-Id: I77d150e860fa344a1604d241e27718150fdb8647
Signed-off-by: Vedashree Vidwans <vvidwans@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-hwpm/+/2982555
Reviewed-by: Adeel Raza <araza@nvidia.com>
Reviewed-by: Vishal Aslot <vaslot@nvidia.com>
GVS: Gerrit_Virtual_Submit <buildbot_gerritrpt@nvidia.com>
Determining whether the header file iosys-map.h is present in the kernel
is currently determine by kernel version. However, for Linux v5.15,
iosys-map.h has been backported in order to support simple-framebuffer
for early display. Therefore, we cannot rely on the kernel version to
indicate whether iosys-map is present. This is also true for 3rd party
Linux kernels that backport changes as well. Fix this by adding a
compile time flag, that will be set accordingly by the conftest script
if this header is present.
Bug 4119327
Bug 4228080
Change-Id: I9de07a4615a6c9da504b36750c48e73e200da301
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-hwpm/+/2974080
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Remove unused vmas parameter from get_user_pages() based
on following change in core kernel.
=====
Upstream commit "54d020692b34"
mm/gup: remove unused vmas parameter from get_user_pages()
Patch series "remove the vmas parameter from GUP APIs", v6.
(pin_/get)_user_pages[_remote]() each provide an optional output parameter
for an array of VMA objects associated with each page in the input range.
These provide the means for VMAs to be returned, as long as mm->mmap_lock
is never released during the GUP operation (i.e. the internal flag
FOLL_UNLOCKABLE is not specified).
In addition, these VMAs can only be accessed with the mmap_lock held and
become invalidated the moment it is released.
The vast majority of invocations do not use this functionality and of
those that do, all but one case retrieve a single VMA to perform checks
upon.
It is not egregious in the single VMA cases to simply replace the
operation with a vma_lookup(). In these cases we duplicate the (fast)
lookup on a slow path already under the mmap_lock, abstracted to a new
get_user_page_vma_remote() inline helper function which also performs
error checking and reference count maintenance.
The special case is io_uring, where io_pin_pages() specifically needs to
assert that the VMAs underlying the range do not result in broken
long-term GUP file-backed mappings.
As GUP now internally asserts that FOLL_LONGTERM mappings are not
file-backed in a broken fashion (i.e. requiring dirty tracking) - as
implemented in "mm/gup: disallow FOLL_LONGTERM GUP-nonfast writing to
file-backed mappings" - this logic is no longer required and so we can
simply remove it altogether from io_uring.
Eliminating the vmas parameter eliminates an entire class of danging
pointer errors that might have occured should the lock have been
incorrectly released.
In addition, the API is simplified and now clearly expresses what it is
intended for - applying the specified GUP flags and (if pinning) returning
pinned pages.
This change additionally opens the door to further potential improvements
in GUP and the possible marrying of disparate code paths.
I have run this series against gup_test with no issues.
Thanks to Matthew Wilcox for suggesting this refactoring!
This patch (of 6):
No invocation of get_user_pages() use the vmas parameter, so remove it.
The GUP API is confusing and caveated. Recent changes have done much to
improve that, however there is more we can do. Exporting vmas is a prime
target as the caller has to be extremely careful to preclude their use
after the mmap_lock has expired or otherwise be left with dangling
pointers.
Removing the vmas parameter focuses the GUP functions upon their primary
purpose - pinning (and outputting) pages as well as performing the actions
implied by the input flags.
This is part of a patch series aiming to remove the vmas parameter
altogether.
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
=====
Bug 4276500
Change-Id: Ie2833b7aa4e8fef1362694de6e8a27bba553e3d4
Signed-off-by: Shardar Mohammed <smohammed@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-hwpm/+/2978634
Reviewed-by: Laxman Dewangan <ldewangan@nvidia.com>
GVS: Gerrit_Virtual_Submit <buildbot_gerritrpt@nvidia.com>
The config flags defined in Kconfig file are not available/used with
OOT kernel builds. To support all kernel versions, HWPM compiles
independent of CONFIG_TEGRA_SOC_HWPM flag. This also applies to
IP config flags which are not supported as well. Hence,
include HWPM IP files irrespective of the IP config flag status.
For OOT builds, use tegra_is_hypervisor_mode() instead of using
static function defined in HWPM driver.
Bug 4061775
Change-Id: Ifab4ad5c7c652a4ad17820a82b363e92280fdd1a
Signed-off-by: Vedashree Vidwans <vvidwans@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-hwpm/+/2918870
Reviewed-by: Vasuki Shankar <vasukis@nvidia.com>
Reviewed-by: Adeel Raza <araza@nvidia.com>
GVS: Gerrit_Virtual_Submit <buildbot_gerritrpt@nvidia.com>
Building HWPM with Linux v5.14 currently fails because the arguments
passed to the dma_buf_vmap and dma_buf_vunmap functions is incorrect.
The arguments passed to the dma_buf_vmap and dma_buf_vunmap functions
were actually updated in Linux kernel version v5.11 and not v5.15. So
correct the kernel version in the HWPM driver so that the correct
version of the APIs are used for Linux kernels v5.11+.
Finally, drop the kernel version check around the inclusion of module.h
and the import of the module namespace because this is not necessary.
Bug 3918909
Change-Id: I98eef5c69af058abfd5ac925614954404371175c
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvidia/+/2860794
(cherry picked from commit 72f00b64c3871e69e7242516849a76bde106ae17)
Reviewed-on: https://git-master.nvidia.com/r/c/linux-hwpm/+/2888555
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
- Create tegra_hwpm_element_enable() instead of directly using
perfmon_enable() HAL. This will allow us to expand
tegra_hwpm_element_enable in future.
- Update log messages in ip structure init code and floorsweep info
function.
- It is possible that IP instances and elements to have 0 start range
address. So, modify check for available elements to use range end
instead.
- Use tegra_hwpm_fake_readl() and tegra_hwpm_fake_writel() macros
instead of fake_readl() and fake_write() functions. That way we have
similar implementation of IO functions and macros can be used across
OSes.
- Check that reserve perfmon function is invoked only for HWPM
components. This check will be useful for expansion in types of
components in the future.
- Clean up and rearrange tegra_hwpm_regops_readl_impl() and
tegra_hwpm_regops_writel_impl() to have designated code corresponding to
the element type.
- Currently, device open and release functions are incorrectly using
clock enable/disable functions instead of using HALs. Correct open and
close functions to use lock HALs.
- Currently, tegra_hwpm_update_mem_bytes() doesn't validate mem_mgmt
structure allocation before accessing mem_bytes_kernel pointer. This can
lead to kernel crash. Update tegra_hwpm_update_mem_bytes() to return
error if mem_mgmt structure s not allocated.
Jira THWPM-74
Change-Id: Ia40bd51187e5ea08572dbee81e577dacf5fb66b6
Signed-off-by: Vedashree Vidwans <vvidwans@nvidia.com>
(cherry picked from commit 411f07484d68dfde0d350a5c67f2748e876b11b8)
Reviewed-on: https://git-master.nvidia.com/r/c/linux-hwpm/+/2888553
Reviewed-by: Adeel Raza <araza@nvidia.com>
GVS: Gerrit_Virtual_Submit <buildbot_gerritrpt@nvidia.com>
Add support to make the HWPM driver as module.
Set appropriate path and environment variable
which needs to be passed when building the driver
as module.
Use basic fuse.h file instead of using fuse-helper.h which is not
available in OOT builds.
Introduce HWPM mock file to add definitions for register and
unregister used if HWPM is not compiled in a build.
Bug 4088571
Change-Id: Id6cfb78356d752b963ad7eb16e39b85df92a9a4b
Signed-off-by: Laxman Dewangan <ldewangan@nvidia.com>
Signed-off-by: Vedashree Vidwans <vvidwans@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-hwpm/+/2894547
GVS: Gerrit_Virtual_Submit <buildbot_gerritrpt@nvidia.com>
Modify OS agnostic files in common, hal and include folder to use MIT
license. This will allow the files to be shared between different OSes.
Modify OS specific files in os/linux and uapi folders to add SPDX
identifier for GPLv2.
Jira THWPM-69
Change-Id: I4fef142354a46fc23b67616204ccf0712a99caec
Signed-off-by: Vedashree Vidwans <vvidwans@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-hwpm/+/2797453
Reviewed-by: Seema Khowala <seemaj@nvidia.com>
GVS: Gerrit_Virtual_Submit <buildbot_gerritrpt@nvidia.com>
HWPM components (PMA, RTR, perfmon) have MMIO address space and a
corresponding virtual address region. It is possible that both MMIO and
virtual addresses are same for an aperture.
MMIO address of an aperture is used in device node to enable the
aperture and further to map HWPM component in the driver.
Virtual addresses are used by the applications to execute regops on HWPM
apertures. Virtual addresses are also used to fake aperture address
space in simulation.
This patch updates
- HWPM aperture structures to include MMIO address.
- aperture ioremap function to use MMIO address values.
- fake register allocation to use virtual address values.
Jira THWPM-41
Change-Id: I05acb68dcb278722cd333e1187b2355d1d739e93
Signed-off-by: Vedashree Vidwans <vvidwans@nvidia.com>
(cherry picked from commit 1c0e8107b4cddad7532c10dddc22bb30cef2540b)
Reviewed-on: https://git-master.nvidia.com/r/c/linux-hwpm/+/2853213
Reviewed-by: Seema Khowala <seemaj@nvidia.com>
GVS: Gerrit_Virtual_Submit <buildbot_gerritrpt@nvidia.com>
