Skip to content

Commit

Permalink
debian: update changelog
Browse files Browse the repository at this point in the history
Signed-off-by: Colin Ian King <colin.i.king@gmail.com>
  • Loading branch information
ColinIanKing committed Apr 4, 2024
1 parent a6336f5 commit 519151f
Showing 1 changed file with 110 additions and 0 deletions.
110 changes: 110 additions & 0 deletions debian/changelog
@@ -1,3 +1,113 @@
stress-ng (0.17.07-1) unstable; urgency=medium

[Stian Onarheim]
* core-stressors.h: Sort the stressor list

[Amit Singh Tomar]
* stress-vm: fix the stress_vm_check argument name

[Erik Stahlman]
* Remove write potentially outside of buffer

[Colin Ian King]
* Makefile: bump version
* Manual: fix table T{ T} around --eigen-method add-double
* core-shared-heap: double size of shared heap allocation slab
* core-helper: move gpu_freq null check to start of function
stress_get_gpu_freq_mhz
* stress-zlib: negative read size check
* core-helper: move null ptr checks earlier in stress_get_meminfo
* core-helper: add #else path in stress_get_load_avg
* stress-fma: voidify return from stress_get_setting call
* core-helper: add endianness information
* stress-pty: break out of loop if EINTR occurs on pty open
* stress-access: set errno to 0 before access calls
* stress-filename: ignore ENOTSUP failures
* stress-msyncmany: break out of test loop if failed more than 5 times
* stress-filename: add ENOSUP error check for Solaris
* stress-clock: ignore EINTR for clock_nanosleep for Solaris
* stress-clock: ignore EINTR for clock_gettime for Solaris
* Add build check and use <sys/mkdev.h> for makedev() macro
* Makefile: enable compile/link garbage collection when GARBAGE_COLLECT
is set to 1
* stress-powmath: increase precision for float tests sum and result to
double
* stress-logmath: increase precision for float tests sum and result to
double
* stress-expmath: increase precision for float tests sum and result to
double
* stress-fma: munmap pfma and not fma
* stress-x86syscall: make function wrap_dummy static
* core-helper: make function stress_fs_magic_to_name static
* stress-powmath: compare floating point values using difference and
precision
* stress-logmath: compare floating point values using difference and
precision
* stress-expmath: compare floating point values using difference and
precision
* stress-enosys: fix system call name __NR_lsm_list_modules
* README.md: update contributor list
* Makefile.config: break functions into 1 line entries and sort them
* Makefile.config: break types into 1 line entries and sort them
* Makefile.config: break cpufeatures into 1 line entries and sort them
* Makefile.config: rename *_WCHAR to *_WCHAR_H to match header naming
* Makefile.config: break headers into 1 line entries and sort them
* core-helper: fix error return code, should be EINVAL
* stress-powmath: remove redundant cast on sum addition
* stress-expmath: remove redundant cast on sum addition
* stress-logmath: remove redundant cast on sum addition
* stress-cgroup: add missing wait for each fork'd child
* core-helper: make stress_fs_magic_to_name static
* core-time: manually inline stress_timespec_to_double into code
* core-perf: make stress_perf_stat_succeeded static
* stress-x86syscall: remove dummy_x86syscalls
* Makefile: add linker garbage collection to remove unreferenced
sections
* core-helper: add some argument sanity checking
* stress-syscall: add maximum system call time in nanoseconds
* stress-syscall: add lsm_list_modules and lsm_get_self_attr system
calls
* stress-enosys: add lsm_list_modules and lsm_get_self_attr system calls
* stress-powmath: add hypotenuse functions
* stress-fma: disable FP_FAST_FMA, FP_FAST_FMAF for now
* stress-mathlog: add logb, logbf and logbl exponent get libc functions
* stress-fma: add the --fma-libc option to use libc fma functions
* README.md: update number of tests
* stress-powmath: add power maths function exercising
* stress-vecshuf: for x86, reduce optimization levels to avoid STLF
* stress-vnni: add build time check for target clone avxvnni
* stress-memhotplug: make variable max non-clobberable by longjmp
* core-target-clones: remove need for gcc version check
* stress-vecshuf: remove OPTIMIZE3 setting
* stress-expmath: add exponential maths function exercising
* README.md: add a couple more paper references
* stress_vm_walking_flush_data: fix buffer overflow on last 7 bytes
* README.md: update contributors list
* stress-logmath: add logarithmic maths function exercising
* README.md: Add another publication link
* memhotplug: add --memhotplug-mmap option
* stress-prio-inv: ensure parent has higher priority than children
* stress-prio-inv: break kill and wait into separate phases
* README.md: add missing newline
* README.md: add some more upstream linux git commit references
* stress-acl: fix typo in unimplemented_reason string
* stress-lsm: exercise new LSM system calls lsm_list_modules and
lsm_get_self_attr
* core-cpu: remove trainling white space
* stress-statmount: Add new stressor to exercise statmount and
listmount system calls
* stress-file-ioct: exercise FS_IOC_GETFSSYSFSPATH
* stress-file-ioctl: add FS_IOC_GETFSSYSFSPATH ioctl
* stress-vm: add walking memory flush stress method
* core-cpu: add a couple more obscure CPU id strings
* stress-monte-carlo: add area of a unit squircle
* kernel-coverage: add --mmap-stressful option
* stress-malloc: add malloc_usable_size allocation size verification
* Makefile.config: filter out -Werror from CFLAGS
* core-perf: remove duplicated "Cgroup Switches" event

-- Colin Ian King <colin.i.king@gmail.com> Thu, 4 Apr 2024 22:33:12 +0100

stress-ng (0.17.06-1) unstable; urgency=medium

[Hsieh-Tseng Shen]
Expand Down

0 comments on commit 519151f

Please sign in to comment.