Bug 1764218
Summary: | glibc: Backport malloc fastbin (mxfast) enhancements | ||
---|---|---|---|
Product: | Red Hat Enterprise Linux 8 | Reporter: | Florian Weimer <fweimer> |
Component: | glibc | Assignee: | Arjun Shankar <ashankar> |
Status: | CLOSED ERRATA | QA Contact: | qe-baseos-tools-bugs |
Severity: | unspecified | Docs Contact: | Oss Tikhomirova <otikhomi> |
Priority: | unspecified | ||
Version: | 8.2 | CC: | ashankar, codonell, dj, fweimer, mnewsome, otikhomi, pfrankli, skolosov |
Target Milestone: | rc | Keywords: | Triaged |
Target Release: | 8.2 | ||
Hardware: | Unspecified | ||
OS: | Unspecified | ||
Whiteboard: | |||
Fixed In Version: | glibc-2.28-81.el8 | Doc Type: | Enhancement |
Doc Text: |
.A new tunable for changing the maximum fastbin size in `glibc`
The `malloc` function uses a series of fastbins that hold reusable memory chunks up to a specific size. The default maximum chunk size is 80 bytes on 32-bit systems and 160 bytes on 64-bit systems. This enhancement introduces a new `glibc.malloc.mxfast` tunable to `glibc` that enables you to change the maximum fastbin size.
|
Story Points: | --- |
Clone Of: | Environment: | ||
Last Closed: | 2020-04-28 16:50:19 UTC | Type: | Bug |
Regression: | --- | Mount Type: | --- |
Documentation: | --- | CRM: | |
Verified Versions: | Category: | --- | |
oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
Cloudforms Team: | --- | Target Upstream Version: | |
Embargoed: | |||
Bug Depends On: | |||
Bug Blocks: | 1746918 |
Description
Florian Weimer
2019-10-22 13:36:29 UTC
These commits also need to be backported in Fedora (30 and later at least). This additional commit was needed in order to fix a bug in set_max_fast that caused a testsuite failure on i386: commit ff12e0fb91b9072800f031cb21fb2651ee7b6251 Author: DJ Delorie <dj> Date: Wed Oct 30 18:03:14 2019 -0400 Base max_fast on alignment, not width, of bins (Bug 24903) set_max_fast sets the "impossibly small" value based on, eventually, MALLOC_ALIGNMENT. The comparisons for the smallest chunk used is, eventually, MIN_CHUNK_SIZE. Note that i386 is the only platform where these are the same, so a smallest chunk *would* be put in a no-fastbins fastbin. This change calculates the "impossibly small" value based on MIN_CHUNK_SIZE instead, so that we can know it will always be impossibly small. Verified with malloc/tst-mxfast Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHSA-2020:1828 |