wireguard-openbsd/sys/uvm, branch jd/queueboosts

Prevent km_alloc() from returning garbage if pagelist is empty.

2020-05-23T06:15:09Z

ok bluhm@, visa@

Document uvmexp.nswget without relying on implementation details.

2020-04-23T07:57:27Z

Prompted by a question from schwarze@ ok deraadt@, schwarze@, visa@

Tweak the code that wakes up uvm_pmalloc sleepers in the page daemin.

2020-04-04T22:08:02Z

Although there are open questions about whether we should flag failures with UVM_PMA_FAIL or not, we really should only wake up a sleeper if we unlink the pma. For now only do that if pages were actually freed in the requested region. Prompted by: CID 1453061 Logically dead code which should be fixed by this commit. ok (and together with) beck@

Do not test against NULL a variable which is dereference before that.

2020-03-25T14:55:14Z

CID 1453116 ok kettenis@

Use FALLTHROUGH in uvm_total() like it is done in uvm_loadav().

2020-03-24T15:03:59Z

CID 1453262.

Do not count pages mapped as PROT_NONE against the RLIMIT_DATA limit.

2020-03-04T21:15:38Z

Instead count (and check the limit) when their protection gets flipped from PROT_NONE to something that permits access. This means that mprotect(2) may now fail if changing the protection would exceed RLIMIT_DATA. This helps code (such as Chromium's JavaScript interpreter that reserves large chunks of address space but populates it sparsely. ok deraadt@, otto@, kurt@, millert@, robert@

Cleanup and includes.

2020-02-18T12:13:39Z

Do not include where it is not needed and stop including in it. ok visa@, anton@

struct vops is not modified during runtime so use const which moves each

2020-01-20T23:21:55Z

into read-only data segment. OK deraadt@ tedu@

Use list for freeing pages in uvn_flush() to optimize freeing chunks of

2020-01-16T15:35:22Z

contiguous pages. ok beck@

Add uvm_anfree_list() to free anons as a list of pages. Use this in

2020-01-04T16:17:29Z

the amap code to free pages as a list instead of one at a time to allow for more efficient freeing. Most of the work done at elk lakes, with testing by me and mlarkin and kettenis. Speeds up a test program which zeros a big pile of memory and then exits considerably. ok kettenis@