summaryrefslogtreecommitdiff
path: root/sys/kern/subr_kmem.c
AgeCommit message (Expand)Author
2023-04-09kmem(9): Tweak branch predictions in fast paths.riastradh
2022-05-30re-do previous - it likely broke kmem cache init.mrg
2022-05-30apply some missing #ifn?def KDTRACE_HOOKS from the previous.mrg
2022-05-30kmem(9): Create dtrace sdt probes for each kmem cache size.riastradh
2022-03-12kmem(9): Show the pointer in kmem_free(..., 0) assertion like before.riastradh
2022-03-12kmem(9): Make kmem_alloc and kmem_free agree about rejecting zero.riastradh
2021-02-06Do not cast memcpy arguments when the intention is unaligned access.joerg
2021-01-24Add kmem_tmpbuf_alloc(), a utility function for allocating memory forthorpej
2020-05-14KASSERT -> panicmaxv
2020-03-08KMEM_SIZE: append the size_t to the allocated buffer, rather thanad
2020-01-25- Pad kmem cache names with zeros so vmstat -m and -C are readable.ad
2019-11-14Add support for Kernel Memory Sanitizer (kMSan). It detects uninitializedmaxv
2019-08-15Retire KMEM_GUARD. It has been superseded by kASan, which is much moremaxv
2019-04-07Provide a code argument in kasan_mark(), and give a code to each caller.maxv
2019-03-26Remove unneeded PR_NOALIGN, pool_allocator_kmem is already page-aligned.maxv
2019-02-04Clobber the size when freeing a buffer. This way, if the same buffer getsmaxv
2018-12-23Simplify the KASAN API, use only kasan_mark() and explain briefly. Themaxv
2018-08-22- opt_kasan.h is included from <sys/asan.h>christos
2018-08-22Reduce the number of KASAN ifdefs, suggested by Christos/Taylor.maxv
2018-08-20Add support for kASan on amd64. Written by me, with some parts inspiredmaxv
2018-08-20Compute the pointer earlier, not in the return statement. No functionalmaxv
2018-08-20Retire KMEM_REDZONE and KMEM_POISON.maxv
2018-01-09add strndup and an alias to strdup.christos
2017-11-09Assert KM_SLEEP xor KM_NOSLEEP in all kmem allocation.riastradh
2017-11-07Add two utility functions to help use kmem with strings: kmem_strdupsize,christos
2017-04-12use opt_kmem.h for the KMEM_ variables.christos
2016-02-29fix vmem_alloc() to never return an error for VM_SLEEP requests,chs
2015-07-27Several changes and improvements in KMEM_GUARD:maxv
2014-07-22Enable KMEM_REDZONE on DIAGNOSTIC. It will try to catch overflows.maxv
2014-07-03Change the pattern of KMEM_REDZONE so that the first byte is never '\0'.maxv
2014-07-02Fix the KMEM_POISON check: it should check the whole buffer, otherwise somemaxv
2014-07-011) Define a malloc(9)-like kmem_header structure for KMEM_SIZE. It is inmaxv
2014-06-251) Make clear that we want the space allocated for the KMEM_SIZE header to bemaxv
2014-06-25Rephrase some comments and remove whitespaces. No functional change.maxv
2014-06-24KMEM_REDZONE+KMEM_POISON is supposed to detect buffer overflows. But it onlymaxv
2014-06-23Enable KMEM_SIZE on DIAGNOSTIC. It will catch memory corruption bugs due to amaxv
2014-06-22Put the KMEM_GUARD code under #if defined(KMEM_GUARD). No functional change.maxv
2013-10-25Mark a diagnostic-only variablemartin
2013-04-22- make debug size check more strictyamt
2013-04-22whitespaceyamt
2013-04-21Whitespace.uebayasi
2013-04-16addresses PR/47512para
2012-07-21split allocation lookup table to decrease overall memory usedpara
2012-04-15We don't support KMEM_GUARD nor FREECHECK yet with rump, so disable themmartin
2012-04-13allow kmem_guard_depth to be set in the config file.mrg
2012-04-01don't overallocated once we leave the cachespara
2012-02-05- Make KMGUARD interrupt-safe.rmind
2012-01-30Fix for KMEM_GUARD; do not use it from interrupt context.rmind
2012-01-28- Instead of kmem_cache_max, calculate max index and avoid a shift.rmind
2012-01-27extending vmem(9) to be able to allocated resources for it's own needs.para