On Fri, Apr 5, 2024 at 6:37 AM Klara Modin <klarasmodin@xxxxxxxxx> wrote:
>
> Hi,
>
> On 2024-03-21 17:36, Suren Baghdasaryan wrote:
> > Overview:
> > Low overhead [1] per-callsite memory allocation profiling. Not just for
> > debug kernels, overhead low enough to be deployed in production.
> >
> > Example output:
> > root@moria-kvm:~# sort -rn /proc/allocinfo
> > 127664128 31168 mm/page_ext.c:270 func:alloc_page_ext
> > 56373248 4737 mm/slub.c:2259 func:alloc_slab_page
> > 14880768 3633 mm/readahead.c:247 func:page_cache_ra_unbounded
> > 14417920 3520 mm/mm_init.c:2530 func:alloc_large_system_hash
> > 13377536 234 block/blk-mq.c:3421 func:blk_mq_alloc_rqs
> > 11718656 2861 mm/filemap.c:1919 func:__filemap_get_folio
> > 9192960 2800 kernel/fork.c:307 func:alloc_thread_stack_node
> > 4206592 4 net/netfilter/nf_conntrack_core.c:2567 func:nf_ct_alloc_hashtable
> > 4136960 1010 drivers/staging/ctagmod/ctagmod.c:20 [ctagmod] func:ctagmod_start
> > 3940352 962 mm/memory.c:4214 func:alloc_anon_folio
> > 2894464 22613 fs/kernfs/dir.c:615 func:__kernfs_new_node
> > ...
> >
> > Since v5 [2]:
> > - Added Reviewed-by and Acked-by, per Vlastimil Babka and Miguel Ojeda
> > - Changed pgalloc_tag_{add|sub} to use number of pages instead of order, per Matthew Wilcox
> > - Changed pgalloc_tag_sub_bytes to pgalloc_tag_sub_pages and adjusted the usage, per Matthew Wilcox
> > - Moved static key check before prepare_slab_obj_exts_hook(), per Vlastimil Babka
> > - Fixed RUST helper, per Miguel Ojeda
> > - Fixed documentation, per Randy Dunlap
> > - Rebased over mm-unstable
> >
> > Usage:
> > kconfig options:
> > - CONFIG_MEM_ALLOC_PROFILING
> > - CONFIG_MEM_ALLOC_PROFILING_ENABLED_BY_DEFAULT
> > - CONFIG_MEM_ALLOC_PROFILING_DEBUG
> > adds warnings for allocations that weren't accounted because of a
> > missing annotation
> >
> > sysctl:
> > /proc/sys/vm/mem_profiling
> >
> > Runtime info:
> > /proc/allocinfo
> >
> > Notes:
> >
> > [1]: Overhead
> > To measure the overhead we are comparing the following configurations:
> > (1) Baseline with CONFIG_MEMCG_KMEM=n
> > (2) Disabled by default (CONFIG_MEM_ALLOC_PROFILING=y &&
> > CONFIG_MEM_ALLOC_PROFILING_BY_DEFAULT=n)
> > (3) Enabled by default (CONFIG_MEM_ALLOC_PROFILING=y &&
> > CONFIG_MEM_ALLOC_PROFILING_BY_DEFAULT=y)
> > (4) Enabled at runtime (CONFIG_MEM_ALLOC_PROFILING=y &&
> > CONFIG_MEM_ALLOC_PROFILING_BY_DEFAULT=n && /proc/sys/vm/mem_profiling=1)
> > (5) Baseline with CONFIG_MEMCG_KMEM=y && allocating with __GFP_ACCOUNT
> > (6) Disabled by default (CONFIG_MEM_ALLOC_PROFILING=y &&
> > CONFIG_MEM_ALLOC_PROFILING_BY_DEFAULT=n) && CONFIG_MEMCG_KMEM=y
> > (7) Enabled by default (CONFIG_MEM_ALLOC_PROFILING=y &&
> > CONFIG_MEM_ALLOC_PROFILING_BY_DEFAULT=y) && CONFIG_MEMCG_KMEM=y
> >
> > Performance overhead:
> > To evaluate performance we implemented an in-kernel test executing
> > multiple get_free_page/free_page and kmalloc/kfree calls with allocation
> > sizes growing from 8 to 240 bytes with CPU frequency set to max and CPU
> > affinity set to a specific CPU to minimize the noise. Below are results
> > from running the test on Ubuntu 22.04.2 LTS with 6.8.0-rc1 kernel on
> > 56 core Intel Xeon:
> >
> > kmalloc pgalloc
> > (1 baseline) 6.764s 16.902s
> > (2 default disabled) 6.793s (+0.43%) 17.007s (+0.62%)
> > (3 default enabled) 7.197s (+6.40%) 23.666s (+40.02%)
> > (4 runtime enabled) 7.405s (+9.48%) 23.901s (+41.41%)
> > (5 memcg) 13.388s (+97.94%) 48.460s (+186.71%)
> > (6 def disabled+memcg) 13.332s (+97.10%) 48.105s (+184.61%)
> > (7 def enabled+memcg) 13.446s (+98.78%) 54.963s (+225.18%)
> >
> > Memory overhead:
> > Kernel size:
> >
> > text data bss dec diff
> > (1) 26515311 18890222 17018880 62424413
> > (2) 26524728 19423818 16740352 62688898 264485
> > (3) 26524724 19423818 16740352 62688894 264481
> > (4) 26524728 19423818 16740352 62688898 264485
> > (5) 26541782 18964374 16957440 62463596 39183
> >
> > Memory consumption on a 56 core Intel CPU with 125GB of memory:
> > Code tags: 192 kB
> > PageExts: 262144 kB (256MB)
> > SlabExts: 9876 kB (9.6MB)
> > PcpuExts: 512 kB (0.5MB)
> >
> > Total overhead is 0.2% of total memory.
> >
> > Benchmarks:
> >
> > Hackbench tests run 100 times:
> > hackbench -s 512 -l 200 -g 15 -f 25 -P
> > baseline disabled profiling enabled profiling
> > avg 0.3543 0.3559 (+0.0016) 0.3566 (+0.0023)
> > stdev 0.0137 0.0188 0.0077
> >
> >
> > hackbench -l 10000
> > baseline disabled profiling enabled profiling
> > avg 6.4218 6.4306 (+0.0088) 6.5077 (+0.0859)
> > stdev 0.0933 0.0286 0.0489
> >
> > stress-ng tests:
> > stress-ng --class memory --seq 4 -t 60
> > stress-ng --class cpu --seq 4 -t 60
> > Results posted at: https://evilpiepirate.org/~kent/memalloc_prof_v4_stress-ng/
> >
> > [2] https://lore.kernel.org/all/20240306182440.2003814-1-surenb@xxxxxxxxxx/
>
> If I enable this, I consistently get percpu allocation failures. I can
> occasionally reproduce it in qemu. I've attached the logs and my config,
> please let me know if there's anything else that could be relevant.
Thanks for the report!
In debug_alloc_profiling.log I see:
[ 7.445127] percpu: limit reached, disable warning
That's probably the reason. I'll take a closer look at the cause of
that and how we can fix it.
In qemu-alloc3.log I see couple of warnings:
[ 1.111620] alloc_tag was not set
[ 1.111880] WARNING: CPU: 0 PID: 164 at
include/linux/alloc_tag.h:118 kfree (./include/linux/alloc_tag.h:118
(discriminator 1) ./include/linux/alloc_tag.h:161 (discriminator 1)
mm/slub.c:2043 ...
[ 1.161710] alloc_tag was not cleared (got tag for fs/squashfs/cache.c:413)
[ 1.162289] WARNING: CPU: 0 PID: 195 at
include/linux/alloc_tag.h:109 kmalloc_trace_noprof
(./include/linux/alloc_tag.h:109 (discriminator 1)
./include/linux/alloc_tag.h:149 (discriminator 1) ...
Which means we missed to instrument some allocation. Can you please
check if disabling CONFIG_MEM_ALLOC_PROFILING_DEBUG fixes QEMU case?
In the meantime I'll try to reproduce and fix this.
Thanks,
Suren.
>
> Kind regards,
> Klara Modin
