user/sven/linux.git/scripts/checkstack.pl, branch v6.19.10

scripts/checkstack.pl: fix no space expression between sp and offset

2023-12-29T20:22:28Z

When I use older version aarch64 objdump (2.24) to disassemble aarch64 vmlinux, I get the result like below. There is no space between sp and offset. ffff800008010000 : ffff800008010000: d503233f hint #0x19 ffff800008010004: a9bc7bfd stp x29, x30, [sp,#-64]! ffff800008010008: 90011e60 adrp x0, ffff80000a3dc000 ffff80000801000c: 910003fd mov x29, sp ffff800008010010: a9025bf5 stp x21, x22, [sp,#32] When I use newer version aarch64 objdump (2.35), I get the result like below. There is a space between sp and offset. ffff800008010000 : ffff800008010000: d503233f paciasp ffff800008010004: a9bc7bfd stp x29, x30, [sp, #-64]! ffff800008010008: 90011e60 adrp x0, ffff80000a3dc000 ffff80000801000c: 910003fd mov x29, sp ffff800008010010: a9025bf5 stp x21, x22, [sp, #32] Add no space support of regular expression for old version objdump. Link: https://lkml.kernel.org/r/20231220073629.2658-1-Kuan-Ying.Lee@mediatek.com Signed-off-by: Kuan-Ying Lee Cc: Casper Li Cc: AngeloGioacchino Del Regno Cc: Chinwen Chang Cc: Matthias Brugger Cc: Qun-Wei Lin Signed-off-by: Andrew Morton

scripts/checkstack.pl: change min_stack to 512 by default

2023-12-29T20:22:27Z

According to Documentation/process/submit-checklist.rst, checkstack does not point out problems explicitly, but any one function that uses more than 512 bytes on the stack is a candidate for change, hence it is better to omit any stack frame sizes smaller than 512 bytes, just change min_stack to 512 by default. Link: https://lkml.kernel.org/r/20231219125008.23007-5-yangtiezhu@loongson.cn Signed-off-by: Tiezhu Yang Cc: Jonathan Corbet Signed-off-by: Andrew Morton

scripts/checkstack.pl: match all stack sizes for some archs

2023-12-29T20:22:27Z

For some unknown reason the regular expression for checkstack only matches three digit numbers starting with the number "3", or any higher number. Which means that it skips any stack sizes smaller than 304 bytes. This makes the checkstack script a bit less useful than it could be. Change the script to match any number. To be filtered out stack sizes can be configured with the min_stack variable, which omits any stack frame sizes smaller than 100 bytes by default. This is similar with commit aab1f809d754 ("scripts/checkstack.pl: match all stack sizes for s390"). Link: https://lkml.kernel.org/r/20231219125008.23007-4-yangtiezhu@loongson.cn Signed-off-by: Tiezhu Yang Cc: Jonathan Corbet Signed-off-by: Andrew Morton

scripts/checkstack.pl: add min_stack to the usage comment

2023-12-29T20:22:26Z

After commit 572220aad525 ("scripts/checkstack.pl: Add argument to print stacks greather than value."), it is appropriate to add min_stack to the usage comment, then the users know explicitly that "min_stack" can be specified like "arch". Link: https://lkml.kernel.org/r/20231219125008.23007-3-yangtiezhu@loongson.cn Signed-off-by: Tiezhu Yang Cc: Jonathan Corbet Signed-off-by: Andrew Morton

scripts/checkstack.pl: remove ia64 support

2023-12-29T20:22:26Z

Patch series "Modify some code about checkstack". This patch (of 5): After commit cf8e8658100d ("arch: Remove Itanium (IA-64) architecture"), the ia64 port has been removed from the kernel, so also remove the ia64 specific bits from the checkstack.pl script. Link: https://lkml.kernel.org/r/20231219125008.23007-1-yangtiezhu@loongson.cn Link: https://lkml.kernel.org/r/20231219125008.23007-2-yangtiezhu@loongson.cn Signed-off-by: Tiezhu Yang Cc: Jonathan Corbet Signed-off-by: Andrew Morton

checkstack: add loongarch support for scripts/checkstack.pl

2023-12-29T20:22:25Z

scripts/checkstack.pl lacks support for the loongarch architecture. Add support to detect "addi.{w,d} $sp, $sp, -FRAME_SIZE" stack frame generation instruction. Link: https://lkml.kernel.org/r/MW4PR84MB314514273F0B7DBCC5E35A978192A@MW4PR84MB3145.NAMPRD84.PROD.OUTLOOK.COM Signed-off-by: Youling Tang Acked-by: Huacai Chen Cc: John Paul Adrian Glaubitz Signed-off-by: Andrew Morton

checkstack: sort output by size and function name

2023-12-11T01:21:32Z

Sort output by size and in addition by function name. This increases readability for cases where there are many functions with the same stack usage. Link: https://lkml.kernel.org/r/20231120183719.2188479-3-hca@linux.ibm.com Signed-off-by: Heiko Carstens Cc: Maninder Singh Cc: Masahiro Yamada Cc: Vaneet Narang Signed-off-by: Andrew Morton

Merge branch 'master' into mm-hotfixes-stable

2023-12-07T01:03:50Z

checkstack: fix printed address

2023-12-07T00:12:46Z

All addresses printed by checkstack have an extra incorrect 0 appended at the end. This was introduced with commit 677f1410e058 ("scripts/checkstack.pl: don't display $dre as different entity"): since then the address is taken from the line which contains the function name, instead of the line which contains stack consumption. E.g. on s390: 0000000000100a30 : ... 100a44: e3 f0 ff 70 ff 71 lay %r15,-144(%r15) So the used regex which matches spaces and hexadecimal numbers to extract an address now matches a different substring. Subsequently replacing spaces with 0 appends a zero at the and, instead of replacing leading spaces. Fix this by using the proper regex, and simplify the code a bit. Link: https://lkml.kernel.org/r/20231120183719.2188479-2-hca@linux.ibm.com Fixes: 677f1410e058 ("scripts/checkstack.pl: don't display $dre as different entity") Signed-off-by: Heiko Carstens Cc: Maninder Singh Cc: Masahiro Yamada Cc: Vaneet Narang Cc: Signed-off-by: Andrew Morton

scripts/checkstack.pl: match all stack sizes for s390

2023-11-22T14:06:23Z

For some unknown reason the regular expression for checkstack only matches three digit numbers starting with the number "3", or any higher number. Which means that it skips any stack sizes smaller than 304 bytes. This makes the checkstack script a bit less useful than it could be. Change the script to match any number. To be filtered out stack sizes can be configured with the min_stack variable, which omits any stack frame sizes smaller than 100 bytes by default. Tested-by: Alexander Gordeev Signed-off-by: Heiko Carstens Signed-off-by: Alexander Gordeev