Fix several typos in the comments

This commit is contained in:
Slobodan Predolac 2026-04-22 09:09:44 -07:00
parent d7b283ed32
commit 6cd31c0985
5 changed files with 7 additions and 7 deletions

View file

@ -195,7 +195,7 @@ fls_u(unsigned x) {
/* \
* Algorithm from an old AMD optimization reference manual. \
* We're putting a little bit more work than you might expect \
* into the no-instrinsic case, since we only support the \
* into the no-intrinsic case, since we only support the \
* GCC intrinsics spelling of popcount (for now). Detecting \
* whether or not the popcount builtin is actually useable in \
* MSVC is nontrivial. \
@ -212,7 +212,7 @@ fls_u(unsigned x) {
* That costs an extra mask relative to this, though. \
*/ \
x = x - ((x >> 1) & (0x55U * bmul)); \
/* Replace each 4 bits with their sideays sum. 0x3 = 0b0011. */\
/* Replace each 4 bits with their sideways sum. 0x3 = 0b0011. */\
x = (x & (bmul * 0x33U)) + ((x >> 2) & (bmul * 0x33U)); \
/* \
* Replace each 8 bits with their sideways sum. Note that we \

View file

@ -145,7 +145,7 @@ struct hpa_shard_s {
bool hpa_hugepage_size_exceeds_limit(void);
/*
* Whether or not the HPA can be used given the current configuration. This is
* Whether or not the HPA can be used given the current configuration. This
* is not necessarily a guarantee that it backs its allocations by hugepages,
* just that it can function properly given the system it's running on.
*/

View file

@ -18,7 +18,7 @@ JEMALLOC_DIAGNOSTIC_DISABLE_SPURIOUS
/* Data. */
/*
* Define names for both unininitialized and initialized phases, so that
* Define names for both uninitialized and initialized phases, so that
* options and mallctl processing are straightforward.
*/
const char *const percpu_arena_mode_names[] = {
@ -44,7 +44,7 @@ static unsigned nbins_total;
/*
* a0 is used to handle huge requests before malloc init completes. After
* that,the huge_arena_ind is updated to point to the actual huge arena,
* that, the huge_arena_ind is updated to point to the actual huge arena,
* which is the last one of the auto arenas.
*/
unsigned huge_arena_ind = 0;

View file

@ -46,7 +46,7 @@ hpa_supported(void) {
return false;
}
/*
* We fundamentally rely on a address-space-hungry growth strategy for
* We fundamentally rely on an address-space-hungry growth strategy for
* hugepages.
*/
if (LG_SIZEOF_PTR != 3) {

View file

@ -43,7 +43,7 @@ const char *je_malloc_conf
* setting, which gets lower priority than the environment settings).
*
* But it's a fairly common use case in some testing environments for a user to
* be able to control the binary, but nothing else (e.g. a performancy canary
* be able to control the binary, but nothing else (e.g. a performance canary
* uses the production OS and environment variables, but can run any binary in
* those circumstances). For these use cases, it's handy to have an in-binary
* mechanism for overriding environment variable settings, with the idea that if