summaryrefslogtreecommitdiff
AgeCommit message (Collapse)AuthorFilesLines
2021-07-02sync usertestsRobert Morris1-0/+11
2021-07-02keep usertests.c in sync with riscv-sol-2020Robert Morris1-0/+27
2020-11-23Rename icache to itableFrans Kaashoek1-25/+25
The inode cache isn't really a cache. The main purpose of it is to allow for synchronization (locking individual inodes), providing long-lived references to inodes, and ensuring that there is only inode in memory.
2020-11-19suppress an incorrect error message in grindRobert Morris1-2/+3
2020-11-05Merge remote-tracking branch 'refs/remotes/origin/riscv' into riscvFrans Kaashoek0-0/+0
2020-11-05xFrans Kaashoek1-1/+1
2020-11-05kill/status testFrans Kaashoek1-0/+31
2020-11-05Be principled: acquire lock firstFrans Kaashoek1-1/+1
2020-11-05proc_tree_lock -> wait_lockRobert Morris1-14/+15
2020-11-05don't over-lock in exit()Robert Morris2-3/+7
2020-11-05xFrans Kaashoek1-1/+1
2020-11-05kill/status testFrans Kaashoek1-0/+31
2020-11-05Be principled: acquire lock firstFrans Kaashoek1-1/+1
2020-11-05proc_tree_lock -> wait_lockRobert Morris1-14/+15
2020-11-05don't over-lock in exit()Robert Morris2-3/+7
2020-11-05oopsRobert Morris1-1/+1
2020-11-05Frans' proc_lock.Robert Morris2-77/+34
2020-11-05free proc if kalloc failsRobert Morris1-0/+1
2020-11-05hopefully make writei more correctRobert Morris2-13/+14
2020-11-05test for closed pipe or killed on every char, not just if pipe fullRobert Morris1-11/+14
2020-11-05fix uart.c to work with UART_TX_BUF_SIZE == 1Robert Morris1-7/+7
2020-11-05consolewrite does not need cons.lock -- can lead to sleep() with lock heldRobert Morris1-2/+0
2020-11-03oopsRobert Morris1-1/+1
2020-11-03Frans' proc_lock.Robert Morris2-77/+34
2020-11-01free proc if kalloc failsRobert Morris1-0/+1
2020-10-23hopefully make writei more correctRobert Morris2-13/+14
2020-10-22test for closed pipe or killed on every char, not just if pipe fullRobert Morris1-11/+14
2020-10-20fix uart.c to work with UART_TX_BUF_SIZE == 1Robert Morris1-7/+7
2020-10-20Merge branch 'riscv' of g.csail.mit.edu:xv6-dev into riscvRobert Morris3-28/+48
2020-10-20consolewrite does not need cons.lock -- can lead to sleep() with lock heldRobert Morris1-2/+0
2020-10-14Fix minor typosFumiya Shigemitsu1-1/+1
2020-10-14kvmmake() makes a complete kernel page table, matching Figure 3.3Frans Kaashoek3-27/+47
2020-10-14set riscv use-compressed-breakpoints yesRobert Morris1-0/+1
2020-10-14Fix minor typosFumiya Shigemitsu1-1/+1
2020-10-14kvmmake() makes a complete kernel page table, matching Figure 3.3Frans Kaashoek3-27/+47
2020-10-07set riscv use-compressed-breakpoints yesRobert Morris1-0/+1
2020-10-05more explicable scratch area size for machine-mode timer interruptsRobert Morris2-12/+12
2020-10-05Don't map the CLINT, since it never used in the kernel in supervisor modeFrans Kaashoek1-3/+0
2020-10-05eliminate virtio DMA into kernel stacks.Robert Morris3-31/+12
2020-10-05improve virtio_disk comments; bring it closer to wording in the specRobert Morris2-37/+85
2020-10-05don't unpin if recovering -- the resulting negative refcnt suppresses next unpinRobert Morris1-4/+5
2020-10-05avoid deadlock by disk intr acking interrupt first, then processing ringRobert Morris2-21/+37
2020-10-05often causes deadlock after a few minutesRobert Morris1-2/+58
2020-10-05more explicable scratch area size for machine-mode timer interruptsRobert Morris2-12/+12
2020-10-05Don't map the CLINT, since it never used in the kernel in supervisor modeFrans Kaashoek1-3/+0
2020-10-05eliminate virtio DMA into kernel stacks.Robert Morris3-31/+12
2020-10-05improve virtio_disk comments; bring it closer to wording in the specRobert Morris2-37/+85
2020-10-04don't unpin if recovering -- the resulting negative refcnt suppresses next unpinRobert Morris1-4/+5
2020-10-04avoid deadlock by disk intr acking interrupt first, then processing ringRobert Morris2-21/+37
2020-10-04often causes deadlock after a few minutesRobert Morris1-2/+58
Powered by cgit, Theme cgit-dark created by Ted Yin, updated by Mole Shang.