On 17/04/2025 13:10, Boris Brezillon wrote:
> On Thu, 17 Apr 2025 11:41:18 +0100
> Steven Price wrote:
>
>> On 17/04/2025 11:05, Boris Brezillon wrote:
>>> Currently, we pick the MMIO offset based on the size of the pgoff_t
>>> type seen by the process that manipulates the FD, such that a 32-bit
On Thu, 17 Apr 2025 14:10:31 +0200
Boris Brezillon wrote:
> > I can't help feeling we can just simplify this to:
> >
> > u64 mmio_offset = pfile->user_mmio.offset;
> >
> > if (offset >= mmio_offset) {
> > offset -= mmio_offset;
> > offset += DRM_PANTHOR_USER_MMIO
On Thu, Apr 17, 2025 at 02:16:11PM +0200, Boris Brezillon wrote:
> On Thu, 17 Apr 2025 12:33:01 +0100
> Liviu Dudau wrote:
>
> > On Thu, Apr 17, 2025 at 11:41:18AM +0100, Steven Price wrote:
> > > On 17/04/2025 11:05, Boris Brezillon wrote:
> > > > Currently, we pick the MMIO offset based on th
On Thu, 17 Apr 2025 12:33:01 +0100
Liviu Dudau wrote:
> On Thu, Apr 17, 2025 at 11:41:18AM +0100, Steven Price wrote:
> > On 17/04/2025 11:05, Boris Brezillon wrote:
> > > Currently, we pick the MMIO offset based on the size of the pgoff_t
> > > type seen by the process that manipulates the FD,
On Thu, 17 Apr 2025 11:41:18 +0100
Steven Price wrote:
> On 17/04/2025 11:05, Boris Brezillon wrote:
> > Currently, we pick the MMIO offset based on the size of the pgoff_t
> > type seen by the process that manipulates the FD, such that a 32-bit
> > process can always map the user MMIO ranges. Bu
On Thu, Apr 17, 2025 at 11:41:18AM +0100, Steven Price wrote:
> On 17/04/2025 11:05, Boris Brezillon wrote:
> > Currently, we pick the MMIO offset based on the size of the pgoff_t
> > type seen by the process that manipulates the FD, such that a 32-bit
> > process can always map the user MMIO range
On 17/04/2025 11:05, Boris Brezillon wrote:
> Currently, we pick the MMIO offset based on the size of the pgoff_t
> type seen by the process that manipulates the FD, such that a 32-bit
> process can always map the user MMIO ranges. But this approach doesn't
> work well for emulators like FEX, where
Currently, we pick the MMIO offset based on the size of the pgoff_t
type seen by the process that manipulates the FD, such that a 32-bit
process can always map the user MMIO ranges. But this approach doesn't
work well for emulators like FEX, where the emulator is a 64-bit binary
which might be exec
