The simulator ring buffer DMA interface supports buffers of the following sizes:
4, 8, 12 and 16K. At present, it is configured to 4K and it happens to match
with the kernel PAGE_SIZE, which is used to wrap back the GET/PUT pointers once
4K is reached. However, this is not always true; for instance, take 64K pages.
Hence, replace PAGE_SIZE with SIM_BFR_SIZE.
Introduce macro NVGPU_CPU_PAGE_SIZE which aliases to PAGE_SIZE and replace
latter with former.
Bug 200658101
Jira NVGPU-6018
Change-Id: I83cc62b87291734015c51f3e5a98173549e065de
Signed-off-by: Antony Clince Alex <aalex@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2420728
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Mapping of large buffers to GMMU end up needing many
pages for the PTE tables. Allocating these one by one
can end up being a performance bottleneck, particularly
in the virtualized case.
This is adding the following changes:
- As the TLB invalidation doesn't have access to mem_off,
allow top-level allocation by alloc_cache_direct().
- Define NVGPU_PD_CACHE_SIZE, the allocation size for a new slab
for the PD cache, effectively set to 64K bytes
- Use the PD cache for any allocation < NVGPU_PD_CACHE_SIZE
When freeing up cached entries, avoid prefetch errors by
invalidating the entry (memset to 0).
- Try to fall back to direct allocation of smaller chunk for
contiguous allocation failures.
- Unit test changes.
Bug 200649243
Change-Id: I0a667af0ba01d9147c703e64fc970880e52a8fbc
Signed-off-by: dt <dt@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2404371
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Add macros for whitelisting coverity violations. These macros use pragma
directives. The pragma directives and whitelisting macros are only
enabled when a coverity scan is being run.
The whitelisting macros have been added to a new header called
static_analysis.h. The contents of safe_ops.h (CERT C safe ops) have
been moved into static_analysis.h because this will be the new header
for static analysis related macros/defines/etc.
JIRA NVGPU-3820
Change-Id: I9c63f20f670880b420415535738034619314b7c3
Signed-off-by: Adeel Raza <araza@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2180600
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Advisory Rule 2.7 states that there should be no unused
parameters in functions.
This patch removes unused function parameters from the following:
* release_as_share_id() -> remove 'id' param
* nvgpu_pd_cache_look_up() -> remove 'g' param
* nvgpu_vm_get_pte_size_fixed_map() -> remove 'size' param
Jira NVGPU-3178
Change-Id: Id2c3b5378bba9bdc0312742fd8393fb3ec67c4df
Signed-off-by: Scott Long <scottl@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2178650
GVS: Gerrit_Virtual_Submit
Reviewed-by: Adeel Raza <araza@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Fix the following MISRA rule violations in bitops unit,
MISRA Rule 10.1
MISRA Rule 10.3
MISRA Rule 10.4
MISRA Rule 11.8
MISRA Rule 21.2
Introduce nvgpu specific functions for bitops and bitmap operations
with unsigned integer as parameter for offset. OS specific type
conversions and handling of these inerfaces are taken care in the
respective OS files.
Jira NVGPU-3545
Change-Id: Ib1ef76563db6ba1d879a0b4d365b2958ea03f85c
Signed-off-by: ajesh <akv@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2129513
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Make the nvgpu_init_mutex function return void.
In linux case, this doesn't affect anything since mutex_init
returns void.
For posix, we assert() and die if pthread_mutex_init fails.
This alleviates the need to error inject for _every_
nvgpu_mutex_init function in the driver.
Jira NVGPU-3476
Change-Id: Ibc801116dc82cdfcedcba2c352785f2640b7d54f
Signed-off-by: Thomas Fleury <tfleury@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2130538
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
The flag NVGPU_DMA_FORCE_CONTIGUOUS simply means that the memory
or the pages should be forced contiguous. Meanwhile, the other
flag NVGPU_DMA_PHYSICALLY_ADDRESSED means that the memory should
be contiguous from GPU perspective, either physically contiguous
when IOMMU is not used, or virtually contiguous by IOMMU.
Thus the NVGPU_DMA_FORCE_CONTIGUOUS flag is now redundant.
This patch cleans up the NVGPU_DMA_FORCE_CONTIGUOUS flag.
Bug 200444660
Change-Id: I63bb06fea728b34ec2c6f831504392d42c426d55
Signed-off-by: Nicolin Chen <nicolinc@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2035403
GVS: Gerrit_Virtual_Submit
Reviewed-by: Alex Waterman <alexw@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
MISRA Rule-17.7 requires the return value of all functions to be used.
Fix is either to use the return value or change the function to return
void. This patch ensures that WARN and WARN_ON always return void; and
introduces a new nvgpu_do_assert construct to trigger the equivalent
of WARN_ON(true) so that stack can be dumped (depends on OS support)
JIRA NVGPU-677
Change-Id: Ie2312c5588ceb5b1db825d15a096149b63b69af4
Signed-off-by: Nicolas Benech <nbenech@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/2018706
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
The pd_cache header declarations were oriignally part of the
gmmu.h header. This is not good from a unit isolation perspective
so this patch moves all the pd_cache specifics over to a new
header file: <nvgpu/pd_cache.h>.
Also a couple of static inlines that were possible when the code
was part of gmmu.h were turned into real, first class functions.
This allowed the pd_cache.h header to not include the gmmu.h
header file.
Also fix an issue in the nvgpu_pd_write() function where the data
was being passed as a size_t for some reason. This has now been
changed to a u32.
JIRA NVGPU-1444
Change-Id: Ib9e9e5a54544de403bfcd8e11c30de05721ddbcc
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1966352
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
This reverts commit 15603b9fd5.
Causes a build break in the PD cache unit test. Not sure how this
passed GVS - must have been a race or something? Unclear.
Change-Id: Ia484a801d098d69441326fa1dd40a1c86e2e23ce
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1966335
The pd_cache header declarations were originally part of the
gmmu.h header. This is not good from a unit isolation perspective
so this patch moves all the pd_cache specifics over to a new
header file: <nvgpu/pd_cache.h>.
Also a couple of static inlines that were possible when the code
was part of gmmu.h were turned into real, first class functions.
This allows the pd_cache.h header to not include the gmmu.h
header file.
Also fix an issue in the nvgpu_pd_write() function where the data
was being passed as a size_t for some reason. This has now been
changed to a u32.
JIRA NVGPU-1444
Change-Id: Iead9a0d998396d2289ffcb3b48765d770400397b
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1965271
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
The problem here, and the solution, requires some background
so let's start there.
During page table programming page directories (PDs) are
allocated as needed. Each PD can range in size, depending on
chip, from 256 bytes all the way up to 32KB (gk20a 2-level
page tables).
In HW, two distinct PTE sizes are supported: large and small.
The HW supports mixing these at will. The second to last level
PDE has pointers to both a small and large PD with
corresponding PTEs. Nvgpu doesn't handle that well and as a
result historically we split the GPU virtual address space
up into a small page region and a large page region. This
makes the GMMU programming logic easier since we now only have
to worry about one type of PD for any given region.
But this presents issues for CUDA and UVM. They want to be
able to mix PTE sizes in the same GPU virtual memory range.
In general we still don't support true dual page directories.
That is page directories with both the small and large next
level PD populated. However, we will allow adjecent PDs to
have different sized next-level PDs.
Each last level PD maps the same amount. On Pascal+ that's
2MB. This is true regardless of the PTE coverage (large or
small). That means the last level PD will be different in
size depending on the PTE size.
So - going back to the SW we allocate PDs as needed when
programming the page tables. When we do this allocation we
allocate just enough space for the PD to contain the
necessary number of PTEs for the page size. The problem
manifests when a PD flips in size from large to small PTEs.
Consider the following mapping operations:
map(gpu_va -> phys) [large-pages]
unmap(gpu_va)
map(gpu_va -> phys) [small-pages]
In the first map/unmap we go and allocate all the necessary
PDs and PTEs to build this translation. We do so assuming a
large page size. When unmapping, as an optimzation/quirk of
nvgpu, we leave the PDs around. We know they may well be used
again in the future.
But if we swap the size of the mapping from large to small
then we now need more space in the PD for PTEs. But the logic
in the GMMU coding assumes if the PD has memory allocated then
that memory is sufficient. This worked back when there was no
potential for a PD to swap in page size. But now that there is
we have to re-allocate the PD doesn't have enough space for
the required PTEs.
So that's the fix - reallocate PDs when they require more
space than they currently have.
Change-Id: I9de70da6acfd20c13d7bdd54232e4d4657840394
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1933076
Reviewed-by: svc-misra-checker <svc-misra-checker@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Nicolas Benech <nbenech@nvidia.com>
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Add two new sub-directories under MM: gmmu and allocators.
The allocators directory is for all the allocator code we have.
There's a fair amount and as such could be considered a component
with a bunch of sub-units.
The new GMMU directory will contain the GMMU component (which used to
be a single unit). The new GMMU component is comprised of the
page_table and pd_cache units. Also when we migrate the chip specific
GMMU code out of mm_gk20a.c and mm_gp10b.c it will be placed in this
new GMMU directory.
JIRA NVGPU-1390
Change-Id: I7aa47ea2a32612b7d69972671fccb72770e1ae09
Signed-off-by: Alex Waterman <alexw@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/1944385
Reviewed-by: Nicolas Benech <nbenech@nvidia.com>
GVS: Gerrit_Virtual_Submit
Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
