-*- text -*- From: "Waqar Mohsin" Subject: 3 questions about switch_threads() in switch.S To: blp@cs.stanford.edu, joshwise@stanford.edu Date: Fri, 3 Mar 2006 17:09:21 -0800 QUESTION 1 In the section # Save current stack pointer to old thread's stack, if any. movl SWITCH_CUR(%esp), %eax test %eax, %eax jz 1f movl %esp, (%eax,%edx,1) 1: # Restore stack pointer from new thread's stack. movl SWITCH_NEXT(%esp), %ecx movl (%ecx,%edx,1), %esp why are we saving the current stack pointer only if the "cur" thread pointer is non-NULL ? Isn't it gauranteed to be non-NULL because switch_threads() is only called form schedule(), where we have struct thread *cur = running_thread (); which should always be non-NULL (given the way kernel pool is laid out). QUESTION 2 # This stack frame must match the one set up by thread_create(). pushl %ebx pushl %ebp pushl %esi pushl %edi I find the comment confusing. thread_create() is a special case: the set of registers popped from switch_threads stack frame for a newly created thread are all zero, so their order shouldn't dictate the order above. I think all that matters is that the order of pops at the end of switch_threads() is the opposite of the pushes at the beginning (as shown above). QUESTION 3 Is it true that struct switch_threads_frame does NOT strictly require struct thread *cur; /* 20: switch_threads()'s CUR argument. */ struct thread *next; /* 24: switch_threads()'s NEXT argument. */ at the end ? When a newly created thread's stack pointer is installed in switch_threads(), all we do is pop the saved registers and return to switch_entry() which pops off and discards the above two simulated (and not used) arguments to switch_threads(). If we remove these two from struct switch_threads_frame and don't do a # Discard switch_threads() arguments. addl $8, %esp in switch_entry(), things should still work. Am I right ? Thanks Waqar From: "Godmar Back" Subject: thread_yield in irq handler To: "Ben Pfaff" Date: Wed, 22 Feb 2006 22:18:50 -0500 Ben, you write in your Tour of Pintos: "Second, an interrupt handler must not call any function that can sleep, which rules out thread_yield(), lock_acquire(), and many others. This is because external interrupts use space on the stack of the kernel thread that was running at the time the interrupt occurred. If the interrupt handler tried to sleep and that thread resumed, then the two uses of the single stack would interfere, which cannot be allowed." Is the last sentence really true? I thought the reason that you couldn't sleep is that you would put effectively a random thread/process to sleep, but I don't think it would cause problems with the kernel stack. After all, it doesn't cause this problem if you call thread_yield at the end of intr_handler(), so why would it cause this problem earlier. As for thread_yield(), my understanding is that the reason it's called at the end is to ensure it's done after the interrupt is acknowledged, which you can't do until the end because Pintos doesn't handle nested interrupts. - Godmar From: "Godmar Back" For reasons I don't currently understand, some of our students seem hesitant to include each thread in a second "all-threads" list and are looking for ways to implement the advanced scheduler without one. Currently, I believe, all tests for the mlfqs are such that all threads are either ready or sleeping in timer_sleep(). This allows for an incorrect implementation in which recent-cpu and priorities are updated only for those threads that are on the alarm list or the ready list. The todo item would be a test where a thread is blocked on a semaphore, lock or condition variable and have its recent_cpu decay to zero, and check that it's scheduled right after the unlock/up/signal. From: "Godmar Back" Subject: set_priority & donation - a TODO item To: "Ben Pfaff" Date: Mon, 20 Feb 2006 22:20:26 -0500 Ben, it seems that there are currently no tests that check the proper behavior of thread_set_priority() when called by a thread that is running under priority donation. The proper behavior, I assume, is to temporarily drop the donation if the set priority is higher, and to reassume the donation should the thread subsequently set its own priority again to a level that's lower than a still active donation. - Godmar From: Godmar Back Subject: project 4 question/comment regarding caching inode data To: Ben Pfaff Date: Sat, 14 Jan 2006 15:59:33 -0500 Ben, in section 6.3.3 in the P4 FAQ, you write: "You can store a pointer to inode data in struct inode, if you want," Should you point out that if they indeed do that, they likely wouldn't be able to support more than 64 open inodes systemwide at any given point in time. (This seems like a rather strong limitation; do your current tests open more than 64 files? It would also point to an obvious way to make the projects harder by specifically disallowing that inode data be locked in memory during the entire time an inode is kept open.) - Godmar From: Godmar Back Subject: on caching in project 4 To: Ben Pfaff Date: Mon, 9 Jan 2006 20:58:01 -0500 here's an idea for future semesters. I'm in the middle of project 4, I've started by implementing a buffer cache and plugging it into the existing filesystem. Along the way I was wondering how we could test the cache. Maybe one could adopt a similar testing strategy as in project 1 for the MLQFS scheduler: add a function that reads "get_cache_accesses()" and a function "get_cache_hits()". Then create a version of pintos that creates access traces for a to-be-determined workload. Run an off-line analysis that would determine how many hits a perfect cache would have (MAX), and how much say an LRU strategy would give (MIN). Then add a fudge factor to account for different index strategies and test that the reported number of cache hits/accesses is within (MIN, MAX) +/- fudge factor. (As an aside - I am curious why you chose to use a clock-style algorithm rather than the more straightforward LRU for your buffer cache implementation in your sample solution. Is there a reason for that? I was curious to see if it made a difference, so I implemented LRU for your cache implementation and ran the test workload of project 4 and printed cache hits/accesses. I found that for that workload, the clock-based algorithm performs almost identical to LRU (within about 1%, but I ran nondeterministally with QEMU). I then reduced the cache size to 32 blocks and found again the same performance, which raises the suspicion that the test workload might not force any cache replacement, so the eviction strategy doesn't matter.) Godmar Back writes: > in your sample solution to P4, dir_reopen does not take any locks when > changing a directory's open_cnt. This looks like a race condition to > me, considering that dir_reopen is called from execute_process without > any filesystem locks held. * Get rid of rox--causes more trouble than it's worth * Reconsider command line arg style--confuses everyone. * Finish writing tour. * Introduce a "yield" system call to speed up the syn-* tests. via Godmar Back: * Project 3 solution needs FS lock. * Get rid of mmap syscall, add sbrk. * Make backtrace program accept multiple object file arguments, e.g. add -u option to allow backtracing user program also. * page-linear, page-shuffle VM tests do not use enough memory to force eviction. Should increase memory consumption. * Add FS persistence test(s). * lock_acquire(), lock_release() don't need additional intr_dis/enable calls, because the semaphore protects lock->holder. * process_death test needs improvement * Internal tests. * Improve automatic interpretation of exception messages. * Userprog project: - Mark read-only pages as actually read-only in the page table. Or, since this was consistently rated as the easiest project by the students, require them to do it. - Don't provide per-process pagedir implementation but only single-process implementation and require students to implement the separation? This project was rated as the easiest after all. Alternately we could just remove the synchronization on pid selection and check that students fix it. * Filesys project: - Need a better way to measure performance improvement of buffer cache. Some students reported that their system was slower with cache--likely, Bochs doesn't simulate a disk with a realistic speed. * Documentation: - Add "Digging Deeper" sections that describe the nitty-gritty x86 details for the benefit of those interested. - Add explanations of what "real" OSes do to give students some perspective. * Assignments: - Add extra credit: . Low-level x86 stuff, like paged page tables. . Specifics on how to implement sbrk, malloc. . Other good ideas. . opendir/readdir/closedir . everything needed for getcwd()