summaryrefslogtreecommitdiff
AgeCommit message (Collapse)AuthorFilesLines
2009-09-02another attempt at cpu-local variables.Russ Cox4-18/+16
this time do it ourselves instead of piggybacking on TLS. add -fno-pic to Makefile; pic code breaks our fake TLS.
2009-09-02lapic: disable microdelay because it makes Bochs slowRuss Cox1-4/+0
2009-09-02fix TLS again;Russ Cox2-3/+7
still not quite but a lot better.
2009-09-02Fix TLS for PIC systemsRuss Cox2-0/+2
2009-09-02fix bootblock build with gcc 4.4Nickolai Zeldovich2-1/+15
2009-08-30assorted fixes:Russ Cox23-2324/+2382
* rename c/cp to cpu/proc * rename cpu.context to cpu.scheduler * fix some comments * formatting for printout
2009-08-08shuffle and tweak for formatting.Russ Cox24-6513/+6710
pdf has very good page breaks now. would be a good copy for fall 2009.
2009-07-13exec tweaksRuss Cox1-6/+11
2009-07-12remove //DOC in the right placeRuss Cox2-1/+1
2009-07-12strip //DOC comments during printingRuss Cox1-0/+1
2009-07-12mergeRuss Cox0-0/+0
2009-07-12more doc tweaksRuss Cox4-54/+52
2009-07-12undoSilas Boyd-Wickizer1-1/+0
2009-07-12test commit emailSilas Boyd-Wickizer1-0/+1
2009-07-11initproc, usegment, swtch tweaksRuss Cox4-60/+46
2009-07-11spinlock: rename parameter lock -> lkRuss Cox1-15/+15
2009-07-11mmu.h: unused SEG_NULLRuss Cox1-3/+0
2009-07-11rearrangements and cleanup for textRuss Cox4-107/+102
2009-07-11trap cleanupRuss Cox3-7/+7
IRQ_OFFSET -> T_IRQ0 (a trap number not an IRQ number) T_SYSCALL 0x30 -> 0x40 (move out of IRQ range)
2009-07-11bootasm cleanupRuss Cox2-12/+12
2009-07-11s/IRQ_OFFSET/T_IRQ0/: it's a trap number, not an irq number.Russ Cox2-9/+10
move the SYSCALL number up, so does not overlap the IRQ traps.
2009-05-31compile fixesrsc1-7/+5
2009-05-31group locks into structs they protect.rsc10-80/+82
few naming nits.
2009-05-31usys.S: put before init.c, STUB -> SYSCALLrsc2-22/+22
2009-05-31exec sanity checkrsc1-1/+1
2009-05-31simplifyrsc3-29/+16
2009-05-31use ftable like btable and itablersc1-22/+21
2009-05-31fix bug fixrsc1-2/+2
2009-05-31tidy fs.c; bmap callers always expected allocationrsc1-24/+15
2009-05-31inode reuse bug.rsc1-1/+1
Suppose an inode has been used and freed. It is left marked I_VALID (the bug). Now ialloc comes along and reuses the inode. It writes the new inode type to disk and returns iget(dev, inum) to get the cache entry. Iget sees that the inode is valid and doesn't bother refreshing from disk. Now when the caller iupdates, it will write out a zero type and the file or directory has disappeared.
2009-05-31bio.c: use struct like icache doesrsc1-30/+31
2009-05-31ide.c: make names more regularrsc1-23/+26
2009-05-31clean up %fs %gs usersc4-8/+5
2009-05-31sync with c; .text is impliedrsc3-18/+25
2009-05-31drop NADDRS and INDIRECT; too many namesrsc4-15/+13
2009-05-31tab policersc4-42/+42
2009-05-31move fork into proc.crsc3-28/+25
2009-05-31missed this file in last checkinrsc1-1/+6
2009-05-31Some proc cleanup, moving some of copyproc into allocproc.rsc9-118/+147
Also, an experiment: use "thread-local" storage for c and cp instead of the #define macro for curproc[cpu()].
2009-05-31move T_DIR etc into stat.h; move type uprsc3-5/+6
2009-05-31Add serial port input/output.rsc4-23/+87
Delete parallel port output. Works well with qemu -nographic mode.
2009-05-31slight rearrangement; sleep on b not buf.rsc1-8/+8
2009-05-26Add MIT license file; update xv6-rev2.tar.gz with LICENSE.rsc3-0/+26
Other files unchanged.
2009-03-08be consistent: no underscores in function namesrsc15-136/+121
2009-03-08xv6: boot loader adjustmentsrsc4-31/+41
do Bochs breakpoint and spin in bootasm.S. not needed in bootmain too. fix readseg bug (rounding of va). zero segments when memsz > filesz. no need to clear BSS in kernel main. make bootother.S like bootasm.S
2009-03-08xv6: use stosb for memsetrsc2-11/+4
2009-03-08xv6/x86.h: add stosb, fix bugs in insl/outsl (rep not repne)rsc1-2/+11
2009-03-08xv6/x86.h: inline assembly cleanuprsc1-8/+8
2008-10-20Explain why itrunc() is OK w.r.t. crashes.rtm1-0/+2
2008-10-17try harder to get directory refcounts rightkolya1-2/+8