kernel stack abuse

Hi!

Nearly five years ago I performed small examination of binary code produced
building 32 bits FreeBSD i386 kernel and found some functions that abused stack
with large structures:

https://lists.freebsd.org/pipermail/svn-src-head/2017-December/107294.html

Now I updated old script for llvm-objdump and ran it for 13.1-STABLE/amd64 GENERIC kernel
and found it went much worse.

The script:

#!/bin/sh
dir=/usr/obj/usr/src/amd64.amd64/sys/GENERIC
objdump=llvm-objdump
set -e

cd $dir
for o in *.o
do
  $objdump -d $o | awk -vn=$o '
    /subq?.*, ?%[er]sp/ {
      split ($(NF-1),a,/[,$]/); printf "%u %s %s\n", a[2], a[2], n }'
done | sort -rn > top.sub

head -50 top.sub | while read d h o
do
  $objdump -d $o | egrep -B8 "subq?.*$h, ?%[er]sp" |\
    awk -vo=$o -vd=$d '/>:$/ {print d, o, $2}'
done > top2.sub
EOF

Results:

33296 fse_decompress.o <FSE_buildDTable>:
21024 fse_decompress.o <FSE_decompress>:
18456 huf_decompress.o <HUF_decompress1X2>:
18456 huf_decompress.o <HUF_decompress4X2>:
18456 huf_decompress.o <HUF_decompress1X2>:
18456 huf_decompress.o <HUF_decompress4X2>:
14352 fse_compress.o <FSE_compress2>:
14352 fse_compress.o <FSE_compress>:
14352 fse_compress.o <FSE_compress2>:
14352 fse_compress.o <FSE_compress>:
10264 huf_decompress.o <HUF_decompress1X1>:
10264 huf_decompress.o <HUF_decompress4X1>:
10264 huf_decompress.o <HUF_decompress1X1>:
10264 huf_decompress.o <HUF_decompress4X1>:
6400 huf_compress.o <HUF_compress1X>:
6400 huf_compress.o <HUF_compress2>:
6400 huf_compress.o <HUF_compress>:
6400 huf_compress.o <HUF_compress1X>:
6400 huf_compress.o <HUF_compress2>:
6400 huf_compress.o <HUF_compress>:
6400 huf_compress.o <HUF_compress1X>:
6400 huf_compress.o <HUF_compress2>:
6400 huf_compress.o <HUF_compress>:
4632 in6_proto.o <icmp6stat_sysctl>:
4352 huf_compress.o <HUF_buildCTable>:
4168 ixl_pf_main.o <ixl_sysctl_dump_debug_data>:
4136 ck_rhs.o <ck_rhs_put_robin_hood>:
4112 fse_compress.o <FSE_buildCTable>:
4104 hist.o <HIST_countFast>:
4096 hist.o <HIST_count>:
3320 in6_proto.o <ip6stat_sysctl>:
2264 md_ddf.o <g_raid_md_ctl_ddf>:
2200 ip6_output.o <ip6_ctloutput>:
2120 ar9300_eeprom.o <ar9300_eeprom_restore_internal_address>:
2104 rt2860.o <rt2860_raw_xmit>:
2088 rt2860.o <rt2860_tx>:
2064 huf_decompress.o <HUF_decompress4X_hufOnly>:
2064 huf_decompress.o <HUF_decompress1X_DCtx>:
2064 huf_decompress.o <HUF_decompress4X_hufOnly>:
2064 huf_decompress.o <HUF_decompress1X_DCtx>:
2056 huf_decompress.o <HUF_decompress1X1_DCtx>:
2056 huf_decompress.o <HUF_decompress1X2_DCtx>:
2056 huf_decompress.o <HUF_decompress4X1_DCtx>:
2056 huf_decompress.o <HUF_decompress4X2_DCtx>:
2056 huf_decompress.o <HUF_decompress4X_DCtx>:
2056 huf_decompress.o <HUF_decompress1X1_DCtx>:
2056 huf_decompress.o <HUF_decompress1X2_DCtx>:
2056 huf_decompress.o <HUF_decompress4X1_DCtx>:
2056 huf_decompress.o <HUF_decompress4X2_DCtx>:
2056 huf_decompress.o <HUF_decompress4X_DCtx>:
2056 huf_decompress.o <HUF_decompress1X1_DCtx>:
2056 huf_decompress.o <HUF_decompress1X2_DCtx>:
2056 huf_decompress.o <HUF_decompress4X1_DCtx>:
2056 huf_decompress.o <HUF_decompress4X2_DCtx>:
2056 huf_decompress.o <HUF_decompress4X_DCtx>:
2056 huf_decompress.o <HUF_decompress1X1_DCtx>:
2056 huf_decompress.o <HUF_decompress1X2_DCtx>:
2056 huf_decompress.o <HUF_decompress4X1_DCtx>:
2056 huf_decompress.o <HUF_decompress4X2_DCtx>:
2056 huf_decompress.o <HUF_decompress4X_DCtx>:
2056 huf_decompress.o <HUF_decompress1X1_DCtx>:
2056 huf_decompress.o <HUF_decompress1X2_DCtx>:
2056 huf_decompress.o <HUF_decompress4X1_DCtx>:
2056 huf_decompress.o <HUF_decompress4X2_DCtx>:
2056 huf_decompress.o <HUF_decompress4X_DCtx>:
2048 huf_decompress.o <HUF_readDTableX1>:
2048 huf_decompress.o <HUF_readDTableX2>:
2048 huf_decompress.o <HUF_readDTableX1>:
2048 huf_decompress.o <HUF_readDTableX2>:
1880 kern_proc.o <kern_proc_out>:
1816 blkback.o <xbb_connect>:
1672 zstd_compress.o <ZSTD_compress>:
1576 fse_compress.o <FSE_compress_wksp>:
1496 scsi_sa.o <saioctl>:
1496 nfs_nfsdserv.o <nfsrvd_rename>:
1480 uipc_shm.o <sysctl_posix_shm_list>:
1448 ar9300_paprd.o <create_pa_curve>:
1432 scsi_enc_ses.o <ses_devids_iter>:
1416 xgbe-sysctl.o <sysctl_coalesce_handler>:
1352 fortuna.o <random_fortuna_pre_read>:

First column shows stack usage in bytes (decimal), then come module name
and function name in question.

For example, sys/contrib/zstd/lib/common/fse_decompress.c, function FSE_buildDTable()
allocates over 32KB on stack. I wonder how it is supposed to run
with default kern.kstack_pages=4 that should be 16KB?