[PATCH v6 0/2] tpm: add support for nonblocking operation
From: Tadeusz Struk
Date: Mon Sep 10 2018 - 13:18:26 EST
The TCG SAPI specification  defines a set of functions, which allow
applications to use the TPM device in either blocking or non-blocking fashion.
Each command defined by the specification has a corresponding
Tss2_Sys_<COMMAND>_Prepare() and Tss2_Sys_<COMMAND>_Complete() call, which
together with Tss2_Sys_ExecuteAsync() is designed to allow asynchronous
mode of operation. Currently the TPM driver supports only blocking calls,
which doesn't allow asynchronous IO operations.
This patch changes it and adds support for nonblocking write and a new poll
function to enable applications, which want to take advantage of this feature.
The new functionality can be tested using standard TPM tools implemented
in , together with modified TCTI from , and an example application
by Philip Tricca . Here is a short description from Philip:
"The example application `glib-tss2-event` uses a glib main event loop
to create an RSA 2048 primary key in the TPM2 NULL hierarchy while
using a glib timer event to time the operation. A GSource object is
used to generate an event when the FD underlying the tss2 function
call has data ready. While the application waits for an event indicating
that the CreatePrimary operation is complete, it counts timer events
that occur every 100ms. Once the CreatePrimary operation completes the
number of timer events that occurred is used to make a rough calculation
of the elapsed time. This value is then printed to the console.
This takes ~300 lines of C code and requires no management or
synchronization of threads. The glib GMainContext is "just a poll()
loop" according to the glib documentation here:
and so supporting 'poll' is the easiest way to integrate with glib /
gtk+. This is true of any other event system that relies on 'poll'
instead of worker threads."
Changes in v6:
- Changed commit message in the first patch to more specific.
- Chenged labels names in tpm-interface.c
Changes in v5:
- Changed the workqueue allocation time back from the first user interface
open to module init.
Changes in v4:
- Changed the way buffer_mutex is handled in nonblocking mode so that
it is not held when write() returns to user space.
Changes in v3:
- Fixed problem reported by 0-dey kbuild test robot around __exitcall.
It complained because there is a module_exit() in another file already.
- Added info on example application from Philip
Changes in v2:
- Split the change into two separate patches. First patch adds a pointer
to the space to the struct file_priv to have access to it from the async job.
This is to avoid memory allocations on every write call. Now everything
what's needed is in the file_priv struct.
- Renamed the 'work' member of the timer to avoid confusion.
Now there are 'timeout_work' and 'async_work'.
- Removed the global wait queue and moved it to file_priv.
- Only creating the work queue when the first file is opened.
Tadeusz Struk (2):
tpm: add ptr to the tpm_space struct to file_priv
tpm: add support for nonblocking operation
drivers/char/tpm/tpm-dev-common.c | 150 +++++++++++++++++++++++++++----------
drivers/char/tpm/tpm-dev.c | 22 +++--
drivers/char/tpm/tpm-dev.h | 19 +++--
drivers/char/tpm/tpm-interface.c | 1
drivers/char/tpm/tpm.h | 1
drivers/char/tpm/tpmrm-dev.c | 31 ++++----
6 files changed, 152 insertions(+), 72 deletions(-)