The code attempts to enable them, but hits an undefined behaviour by
shifting by the entire register's width:

    int lines = 32;
    u32 mask = (1 << lines) - 1;    // 00000000 on x86
    u32 mask = (1 << lines) - 1;    // ffffffff on arm (32)
    u32 mask = (1 << lines) - 1;    // 00000000 on arm64
    u32 mask = (1ULL << lines) - 1; // ffffffff everywhere
Signed-off-by: NAdam Borowski <kilobyte@angband.pl>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Forked from GP106 implementation.

Differences:
- 1 PPC/GPC
- Slightly different grctx magics
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NAlexandre Courbot <acourbot@nvidia.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

GP10B's power is managed by generic PM domains, so it does not require a
VDD regulator. Add this option into the chip function structure.
Signed-off-by: NAlexandre Courbot <acourbot@nvidia.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

GR is similar to GP100, with a few unavailable registers.
Signed-off-by: NAlexandre Courbot <acourbot@nvidia.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

GP10B requires a specific initialization sequence due to the absence of
devinit.
Signed-off-by: NAlexandre Courbot <acourbot@nvidia.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

GP10B's MC is compatible with GP100's, but engines need to be explicitly
put out of ELPG during init.
Signed-off-by: NAlexandre Courbot <acourbot@nvidia.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

GP10B's FB is largely compatible with the GP100 implementation.
Signed-off-by: NAlexandre Courbot <acourbot@nvidia.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

GP10B's FIFO is similar to GP100's, but only allows 512 channels.
Signed-off-by: NAlexandre Courbot <acourbot@nvidia.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

The GP10B firmware is very close to GM20B's. The only difference is that
it supports booting multiple falcons. In order to avoid having too much
functions and structures shared, implement its support in the same
source file as GM20B firmware.
Signed-off-by: NAlexandre Courbot <acourbot@nvidia.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

GP10B's secboot is largely similar to GM20B's. Only differences are MC
base address and the fact that GPCCS is also securely managed.
Signed-off-by: NAlexandre Courbot <acourbot@nvidia.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Allow the MC base address to be specified as an argument for the WPR
region reading function. GP10B uses a different address layout as GM20B,
so this is necessary. Also export the function to be used by GP10B.
Signed-off-by: NAlexandre Courbot <acourbot@nvidia.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

The LS firmware post-run hook is the right place to start said LS
firmware. Moving it here also allows to remove special handling in the
ACR code.
Signed-off-by: NAlexandre Courbot <acourbot@nvidia.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

A LS post-run hook can meet an error meaning the failure of secure boot.
Make sure this can be reported.
Signed-off-by: NAlexandre Courbot <acourbot@nvidia.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Having access to the secboot instance loading a LS firmware can be
useful to LS firmware handlers. At least more useful than just having an
out-of-context subdev pointer.

GP10B's firmware will also need to know the WPR address, which can be
obtained from the secboot instance.
Signed-off-by: NAlexandre Courbot <acourbot@nvidia.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Change the secboot and msgqueue interfaces to take a mask of falcons to
reset instead of a single falcon. The GP10B firmware interface requires
FECS and GPCCS to be booted in a single firmware command.

For firmwares that only support single falcon boot, it is trivial to
loop over the mask and boot each falcons individually.
Signed-off-by: NAlexandre Courbot <acourbot@nvidia.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

The gk20a implementation of instance memory uses vmap()/vunmap() to map
memory regions into the kernel's virtual address space. These functions
may sleep, so protecting them by a spin lock is not safe. This triggers
a warning if the DEBUG_ATOMIC_SLEEP Kconfig option is enabled. Fix this
by using a mutex instead.
Signed-off-by: NThierry Reding <treding@nvidia.com>
Reviewed-by: NAlexandre Courbot <acourbot@nvidia.com>
Tested-by: NAlexandre Courbot <acourbot@nvidia.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Forked from GP106 implementation.

Split out from commit enabling secboot/gr support so that it can be
added to earlier kernels.

Cc: stable@vger.kernel.org	[4.10+]
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

The NV4A (aka NV44A) is an oddity in the family. It only comes in AGP
and PCI varieties, rather than a core PCIE chip with a bridge for
AGP/PCI as necessary. As a result, it appears that the MMU is also
non-functional. For AGP cards, the vast majority of the NV4A lineup,
this worked out since we force AGP cards to use the nv04 mmu. However
for PCI variants, this did not work.

Switching to the NV04 MMU makes it work like a charm. Thanks to mwk for
the suggestion. This should be a no-op for NV4A AGP boards, as they were
using it already.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=70388Signed-off-by: NIlia Mirkin <imirkin@alum.mit.edu>
Cc: stable@vger.kernel.org
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NIlia Mirkin <imirkin@alum.mit.edu>
Fixes: 590801c1 ("drm/nouveau/mpeg: remove dependence on namedb/engctx lookup")
Cc: stable@vger.kernel.org # v4.3+
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

The msgqueue pointer validity should be checked by its owner, not by the
msgqueue code itself to avoid this situation.
Signed-off-by: NAlexandre Courbot <acourbot@nvidia.com>
Reported-by: NJulia Lawall <julia.lawall@lip6.fr>
Signed-off-by: NDave Airlie <airlied@redhat.com>

We were returning PTR_ERR() on a NULL pointer, which obviously won't
work. nvkm_engine_ref() will return an error in case something went
wrong.
Signed-off-by: NAlexandre Courbot <acourbot@nvidia.com>
Reported-by: NDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NDave Airlie <airlied@redhat.com>

This commit reworks the RAM detection algorithm, using RAM-per-LTC to
determine whether a board has a mixed-memory configuration instead of
using RAM-per-FBPA.  I'm not certain the algorithm is perfect, but it
should handle all currently known configurations in the very least.

This should fix GTX 970 boards with 4GiB of RAM where the last 512MiB
isn't fully accessible, as well as only detecting half the VRAM on
GF108 boards.

As a nice side-effect, GP10x memory detection now reuses the majority
of the code from earlier chipsets.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

GF108/GM107 implementations will want slightly different functions for
the upcoming RAM detection improvements.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

This value represents the actual number of bytes recieved on the AUX
channel as the result of a read transaction.
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Apparently sinks are allows to respond with ACK even if they didn't
fully complete a transaction...  It seems like a missed opportunity
for DEFER to me, but what do I know :)
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

All the bricks are in place for secure boot to be enabled. This in turn
makes GR usable so enable them all.
Signed-off-by: NAlexandre Courbot <acourbot@nvidia.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Differences from GP100:
- 3 PPCs/GPC.
- Another random reg to calculate/write.
- Attrib CB setup a little different.
- PascalB
- PascalComputeB
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Add support for the msgqueue firmware used to process SEC2 commands
for gp10x chips.
Signed-off-by: NAlexandre Courbot <acourbot@nvidia.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

These gp10x chips are supporting using (roughly) the same firmware.
Compared to previous secure chips, ACR runs on SEC2 and so does the
low-secure msgqueue.

ACR for these chips is based on r367.
Signed-off-by: NAlexandre Courbot <acourbot@nvidia.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

We will also need to load HS blobs outside of acr_r352 (for instance, to
run the NVDEC VPR scrubber), so make this code reusable.
Signed-off-by: NAlexandre Courbot <acourbot@nvidia.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

r375 ACR uses a unified bootloader descriptor for the GR and PMU
firmwares.
Signed-off-by: NAlexandre Courbot <acourbot@nvidia.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

r367 uses a different hsflcn_desc layout and LS firmware signature
format, requiring a rewrite of some functions.

It also makes use of the shadow region, and uses SEC as the boot falcon.
Signed-off-by: NAlexandre Courbot <acourbot@nvidia.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

r364 is similar to r361, but uses a different hsflcn_desc structure to
introduce the shadow region address (even though it is not yet used by
this version).
Signed-off-by: NAlexandre Courbot <acourbot@nvidia.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

For some unknown reason the LS SEC2 firmware needs to be started twice
to operate. Detect and address that condition.
Signed-off-by: NAlexandre Courbot <acourbot@nvidia.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

I had the brilliant idea to "improve" the binary format by removing
a useless indirection in the HS binary files. In the end it just
makes things more complicated than they ought to be as NVIDIA-provided
files need to be adapted. Since the format used can be identified by the
header, support both.
Signed-off-by: NAlexandre Courbot <acourbot@nvidia.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

If the load and unload falcons are different, then a different
bootloader must also be used. Support this case.
Signed-off-by: NAlexandre Courbot <acourbot@nvidia.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>

Since the HS blobs are provided and signed by NVIDIA, we cannot expect
always-consistent behavior. In this case, on GP10x the unload blob may
return 0x1d even though things have run perfectly well. This behavior
has been confirmed by NVIDIA.

So let the callers of the run_blob() hook receive the blob return's
value (a positive integer) and decide what it means. This allows us to
workaround the 0x1d code instead of issuing an error.
Signed-off-by: NAlexandre Courbot <acourbot@nvidia.com>
Signed-off-by: NBen Skeggs <bskeggs@redhat.com>