Skip to content

K
Kernel

openeuler / Kernel
大约 2 年 前同步成功

通知 8
Star 0
Fork 0
K
Kernel
切换分支/标签
  1. 29 1月, 2020 2 次提交
  3. 22 1月, 2020 1 次提交
    • M
      kbuild: use -S instead of -E for precise cc-option test in Kconfig · 3bed1b7b
      由 Masahiro Yamada 提交于 
      Currently, -E (stop after the preprocessing stage) is used to check
whether the given compiler flag is supported.

While it is faster than -S (or -c), it can be false-positive. You need
to run the compilation proper to check the flag more precisely.

For example, -E and -S disagree about the support of
"--param asan-instrument-allocas=1".

$ gcc -Werror --param asan-instrument-allocas=1 -E -x c /dev/null -o /dev/null
$ echo $?
0

$ gcc -Werror --param asan-instrument-allocas=1 -S -x c /dev/null -o /dev/null
cc1: error: invalid --param name ‘asan-instrument-allocas’; did you mean ‘asan-instrument-writes’?
$ echo $?
1
Signed-off-by: NMasahiro Yamada <masahiroy@kernel.org>
      3bed1b7b
  5. 21 1月, 2020 1 次提交
  7. 16 1月, 2020 1 次提交
  9. 15 1月, 2020 2 次提交
    • M
      kbuild: remove *.tmp file when filechk fails · 88fe89a4
      由 Masahiro Yamada 提交于 
      Bartosz Golaszewski reports that when "make {menu,n,g,x}config" fails
due to missing packages, a temporary file is left over, which is not
ignored by git.

For example, if GTK+ is not installed:

  $ make gconfig
  *
  * Unable to find the GTK+ installation. Please make sure that
  * the GTK+ 2.0 development package is correctly installed.
  * You need gtk+-2.0 gmodule-2.0 libglade-2.0
  *
  scripts/kconfig/Makefile:208: recipe for target 'scripts/kconfig/gconf-cfg' failed
  make[1]: *** [scripts/kconfig/gconf-cfg] Error 1
  Makefile:567: recipe for target 'gconfig' failed
  make: *** [gconfig] Error 2
  $ git status
  HEAD detached at v5.4
  Untracked files:
    (use "git add <file>..." to include in what will be committed)

          scripts/kconfig/gconf-cfg.tmp

  nothing added to commit but untracked files present (use "git add" to track)

This is because the check scripts are run with filechk, which misses
to clean up the temporary file on failure.

When the line

  { $(filechk_$(1)); } > $@.tmp;

... fails, it exits immediately due to the 'set -e'. Use trap to make
sure to delete the temporary file on exit.

For extra safety, I replaced $@.tmp with $(dot-target).tmp to make it
a hidden file.
Reported-by: NBartosz Golaszewski <bgolaszewski@baylibre.com>
Signed-off-by: NMasahiro Yamada <masahiroy@kernel.org>
      88fe89a4
    • M
      modpost: assume STT_SPARC_REGISTER is defined · 5370d4ac
      由 Masahiro Yamada 提交于 
      Commit 8d529014 ("[SPARC]: Deal with glibc changing macro names in
modpost.c") was more than 14 years ago. STT_SPARC_REGISTER is hopefully
defined in elf.h of recent C libraries.
Signed-off-by: NMasahiro Yamada <masahiroy@kernel.org>
Acked-by: NDavid S. Miller <davem@davemloft.net>
      5370d4ac
  11. 14 1月, 2020 1 次提交
  13. 07 1月, 2020 5 次提交
    • M
      kbuild: create modules.builtin without Makefile.modbuiltin or tristate.conf · 8b41fc44
      由 Masahiro Yamada 提交于 
      Commit bc081dd6 ("kbuild: generate modules.builtin") added
infrastructure to generate modules.builtin, the list of all
builtin modules.

Basically, it works like this:

  - Kconfig generates include/config/tristate.conf, the list of
    tristate CONFIG options with a value in a capital letter.

  - scripts/Makefile.modbuiltin makes Kbuild descend into
    directories to collect the information of builtin modules.

I am not a big fan of it because Kbuild ends up with traversing
the source tree twice.

I am not sure how perfectly it should work, but this approach cannot
avoid false positives; even if the relevant CONFIG option is tristate,
some Makefiles forces obj-m to obj-y.

Some examples are:

  arch/powerpc/platforms/powermac/Makefile:
    obj-$(CONFIG_NVRAM:m=y)         += nvram.o

  net/ipv6/Makefile:
    obj-$(subst m,y,$(CONFIG_IPV6)) += inet6_hashtables.o

  net/netlabel/Makefile:
    obj-$(subst m,y,$(CONFIG_IPV6)) += netlabel_calipso.o

Nobody has complained about (or noticed) it, so it is probably fine to
have false positives in modules.builtin.

This commit simplifies the implementation. Let's exploit the fact
that every module has MODULE_LICENSE(). (modpost shows a warning if
MODULE_LICENSE is missing. If so, 0-day bot would already have blocked
such a module.)

I added MODULE_FILE to <linux/module.h>. When the code is being compiled
as builtin, it will be filled with the file path of the module, and
collected into modules.builtin.info. Then, scripts/link-vmlinux.sh
extracts the list of builtin modules out of it.

This new approach fixes the false-positives above, but adds another
type of false-positives; non-modular code may have MODULE_LICENSE()
by mistake. This is not a big deal, it is just the code is always
orphan. We can clean it up if we like. You can see cleanup examples by:

  $ git log --grep='make.* explicitly non-modular'

To sum up, this commits deletes lots of code, but still produces almost
equivalent results. Please note it does not increase the vmlinux size at
all. As you can see in include/asm-generic/vmlinux.lds.h, the .modinfo
section is discarded in the link stage.
Signed-off-by: NMasahiro Yamada <masahiroy@kernel.org>
      8b41fc44
    • M
      kbuild: pass KBUILD_MODFILE when compiling builtin objects · 1664a377
      由 Masahiro Yamada 提交于 
      When compiling, Kbuild passes KBUILD_BASENAME (basename of the object)
and KBUILD_MODNAME (basename of the module).

This commit adds another one, KBUILD_MODFILE, which is the path of
the module. (or, the path of the module it would end up in if it were
compiled as a module.)

The next commit will use this to generate modules.builtin without
tristate.conf.
Signed-off-by: NMasahiro Yamada <masahiroy@kernel.org>
      1664a377
    • M
      kbuild: add stringify helper to quote a string passed to C files · 7e826c44
      由 Masahiro Yamada 提交于 
      Make $(squote)$(quote)...$(quote)$(squote) a helper macro.
I will reuse it in the next commit.
Signed-off-by: NMasahiro Yamada <masahiroy@kernel.org>
      7e826c44
    • M
      kbuild: use pattern rule for building built-in.a in sub-directories · a7499267
      由 Masahiro Yamada 提交于 
      The built-in.a in a sub-directory is created by descending into that
directory. It does not depend on the other sub-directories. Loosen
the dependency.
Signed-off-by: NMasahiro Yamada <masahiroy@kernel.org>
      a7499267
    • M
      kbuild: do not create orphan built-in.a or obj-y objects · 56d58936
      由 Masahiro Yamada 提交于 
      Both 'obj-y += foo/' and 'obj-m += foo/' request Kbuild to visit the
sub-directory foo/, but the difference is that only the former combines
foo/built-in.a into the built-in.a of the current directory because
everything in sub-directories visited by obj-m is supposed to be modular.

So, it makes sense to create built-in.a only if that sub-directory is
reachable by the chain of obj-y. Otherwise, built-in.a will not be
linked into vmlinux anyway. For the same reason, it is pointless to
compile obj-y objects in the directory visited by obj-m.
Signed-off-by: NMasahiro Yamada <masahiroy@kernel.org>
      56d58936
  15. 03 1月, 2020 2 次提交
  17. 17 12月, 2019 2 次提交
  19. 14 12月, 2019 3 次提交
    • M
      scripts/kallsyms: fix offset overflow of kallsyms_relative_base · fd2ab2f6
      由 Masahiro Yamada 提交于 
      Since commit 5e5c4fa7 ("scripts/kallsyms: shrink table before
sorting it"), kallsyms_relative_base can be larger than _text, which
causes overflow when building the 32-bit kernel.

https://lkml.org/lkml/2019/12/7/156

This is because _text is, unless --all-symbols is specified, now
trimmed from the symbol table before record_relative_base() is called.

Handle the offset signedness also for kallsyms_relative_base. Introduce
a new helper, output_address(), to reduce the code duplication.

Fixes: 5e5c4fa7 ("scripts/kallsyms: shrink table before sorting it")
Reported-by: NOlof Johansson <olof@lixom.net>
Signed-off-by: NMasahiro Yamada <masahiroy@kernel.org>
      fd2ab2f6
    • M
      mkcompile_h: use printf for LINUX_COMPILE_BY · c8f3dea9
      由 Masahiro Yamada 提交于 
      Commit 858805b3 ("kbuild: add $(BASH) to run scripts with
bash-extension") shed light on portability issues. Here is another one.

Since commit f0772604 ("Fix handling of backlash character in
LINUX_COMPILE_BY name"), we must escape a backslash contained in
LINUX_COMPILE_BY. This is not working on such distros as Ubuntu.

As the POSIX spec [1] says, if any of the operands contain a backslash
( '\' ) character, the results are implementation-defined.

The actual shell of /bin/sh could be bash, dash, etc. depending on
distros, and the behavior of builtin echo command is different among
them.

The bash builtin echo, unless -e is given, copies the arguments to
stdout without expanding escape sequences (BSD-like behavior).

The dash builtin echo, in contrast, adopts System V behavior, which
does expand escape sequences without any option given.

Even non-builtin /bin/echo behaves differently depending on the system.
Due to these variations, echo is considered as a non-portable command.
Using printf is the common solution to avoid the portability issue.

[1] https://pubs.opengroup.org/onlinepubs/009695399/utilities/echo.html

Fixes: 858805b3 ("kbuild: add $(BASH) to run scripts with bash-extension")
Reported-by: NXXing Wei <xxing.wei@unisoc.com>
Signed-off-by: NMasahiro Yamada <masahiroy@kernel.org>
      c8f3dea9
    • M
      mkcompile_h: git rid of UTS_TRUNCATE from LINUX_COMPILE_{BY,HOST} · e8193650
      由 Masahiro Yamada 提交于 
      UTS_VERSION is set to struct uts_namespace, hence a too long string
should be truncated so it fits in 64 characters.

On the other hand, LINUX_COMPILE_BY/HOST are not set to uts_namespace.
They are just used in the banners, which do not have specific length
limitation.

I dug into the git history, but I could not find the reason why
these two strings must fit in 64 characters. Remove them.

Now that UTS_VERSION is the only user of UTS_TRUNCATE, I squashed it.
Signed-off-by: NMasahiro Yamada <masahiroy@kernel.org>
      e8193650
  21. 06 12月, 2019 1 次提交
  23. 05 12月, 2019 3 次提交
  25. 01 12月, 2019 1 次提交
  27. 29 11月, 2019 1 次提交
    • S
      bpf: Force .BTF section start to zero when dumping from vmlinux · df786c9b
      由 Stanislav Fomichev 提交于 
      While trying to figure out why fentry_fexit selftest doesn't pass for me
(old pahole, broken BTF), I found out that my latest patch can break vmlinux
.BTF generation. objcopy preserves section start when doing --only-section,
so there is a chance (depending on where pahole inserts .BTF section) to
have leading empty zeroes. Let's explicitly force section offset to zero.

Before:

$ objcopy --set-section-flags .BTF=alloc -O binary \
	--only-section=.BTF vmlinux .btf.vmlinux.bin
$ xxd .btf.vmlinux.bin | head -n1
00000000: 0000 0000 0000 0000 0000 0000 0000 0000  ................

After:

$ objcopy --change-section-address .BTF=0 \
	--set-section-flags .BTF=alloc -O binary \
	--only-section=.BTF vmlinux .btf.vmlinux.bin
$ xxd .btf.vmlinux.bin | head -n1
00000000: 9feb 0100 1800 0000 0000 0000 80e1 1c00  ................
          ^BTF magic

As part of this change, I'm also dropping '2>/dev/null' from objcopy
invocation to be able to catch possible other issues (objcopy doesn't
produce any warnings for me anymore, it did before with --dump-section).

Fixes: da5fb182 ("bpf: Support pre-2.25-binutils objcopy for vmlinux BTF")
Signed-off-by: NStanislav Fomichev <sdf@google.com>
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>
Acked-by: NJohn Fastabend <john.fastabend@gmail.com>
Cc: Andrii Nakryiko <andriin@fb.com>
Link: https://lore.kernel.org/bpf/20191127225759.39923-1-sdf@google.com
      df786c9b
  29. 28 11月, 2019 1 次提交
  31. 25 11月, 2019 13 次提交