Changes between Version 89 and Version 90 of GSoC/2020

Timestamp:

08/12/20 15:05:02 (4 years ago)

Author:

UTKARSH RAI

Comment:

--

GSoC/2020

@@ -276 +276 @@
            * Complete the test for stack-isolation and sharing by using the high-level APIs
 
-
-     
-
+* '''July 29:''' Ninth Meeting
+      * ''Progress:''
+           * I spent a major part of the week in resolving the problems in changing memory entries of an allocated stack,
+             finally, I realized that the fatal exceptions that I was getting was due to the fact that stacks allocated from
+             the workspace is not necessarily page-aligned and changing their attributes causes exceptions.
+           * Once I resolved this, I worked on improving the previous patch.
+           * I was able to write tests for stack isolating and stack-sharing. After I have resolved all the formatting problems,
+             I will be sending in my improved patches along with the tests for review today.
+      * ''Blockers:''
+           *  None as of yet.
+      * ''Goals for this week:''
+           * Work on providing a better configuration scheme for thread_stack protection. Right now, we have to define (USE_THREAD_STACK_PROTECTION)
+             across multiple files.
+           * Work on my blog and detail all that has to be done for sharing protected stacks.
+
+* '''August 5:''' Tenth Meeting
+      * ''Progress:''
+           * I was finally able to have a stable working solution for stack isolation and stack sharing.
+           * Worked on the design problems in my patch, integrated the stack protection attributes to the existing structures.
+             I will be sending v3 today.
+           * Tried out the new build system and tried to understand its internals for resolving the configuration issues.
+      * ''Blockers:''
+           * Some of the internals of the build system, especially how to use the build specification items to configure isolated
+             threads is still not clear to me. I will give it another go before asking for help on ml.
+      * ''Goals for this week:''
+           * My primary objective this week is to decide and work on the configuration option.
+           * If this is resolved early, I would start working on providing support for thread_stack isolating for an MPU based system (RISC-V).
+
+* '''August 12:''' Eleventh Meeting
+      * ''Progress:''
+           * Decided on the configuration mechanism and implemented it using the new build system.
+           * Resolved all of the issues with the high-level design and tested it with the new build system configuration option.
+           * Looked into the work that needs to be done for providing RISC-V based MPU suppport.
+      * ''Blockers:''
+           * There are a few limitations in my design -
+               * Whenever context-switch takes place from an interrupt, I get fatal errors.
+               * At times when the thread acquires a mutex for a critical section I get INTERNAL_ERROR_BAD_THREAD_DISPATCH_ENVIRONMENT fatal code.
+                 I will discuss more about this on the mailing list.
+      * ''Goals for this week:''
+           * Blocker resolution.
+           * Implement static initialization of the RISC-V MPU( PMP ) system.