| Message ID | 1492149003-19136-2-git-send-email-yamada.masahiro@socionext.com |
|---|---|
| State | Superseded |
| Headers | show |
| Series | kbuild: cleanup asm-offset generation, and make it work with clang | expand |
El Fri, Apr 14, 2017 at 02:50:02PM +0900 Masahiro Yamada ha dit: > This part ended up in redundant code after touched by multiple > people. > > [1] Commit 3234282f33b2 ("x86, asm: Fix CFI macro invocations to > deal with shortcomings in gas") added parentheses for defined > expressions to support old gas for x86. > > [2] Commit a22dcdb0032c ("x86, asm: Fix ancient-GAS workaround") > split the pattern into two to avoid parentheses for non-numeric > expressions. > > [3] Commit 95a2f6f72d37 ("Partially revert patch that encloses > asm-offset.h numbers in brackets") removed parentheses from numeric > expressions as well because parentheses in MN10300 assembly have a > special meaning (pointer access). > > Apparently, there is a conflict between [1] and [3]. After all, > [3] took precedence, and a long time has passed since then. > > Now, merge the two patterns again because the first one is covered > by the other. > > Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com> Reviewed-by: Matthias Kaehlcke <mka@chromium.org> Thanks Matthias
Masahiro Yamada <yamada.masahiro@socionext.com> wrote: > This part ended up in redundant code after touched by multiple > people. > > [1] Commit 3234282f33b2 ("x86, asm: Fix CFI macro invocations to > deal with shortcomings in gas") added parentheses for defined > expressions to support old gas for x86. > > [2] Commit a22dcdb0032c ("x86, asm: Fix ancient-GAS workaround") > split the pattern into two to avoid parentheses for non-numeric > expressions. > > [3] Commit 95a2f6f72d37 ("Partially revert patch that encloses > asm-offset.h numbers in brackets") removed parentheses from numeric > expressions as well because parentheses in MN10300 assembly have a > special meaning (pointer access). > > Apparently, there is a conflict between [1] and [3]. After all, > [3] took precedence, and a long time has passed since then. There's a conflict between [1] and various assembly code formats. Some formats define, say, mov 4,r1 to move the number 4 into register r1, and: mov (4),r1 to move the contents of the memory at address 4 into r1. Therefore, you cannot simply wrap numeric operands in brackets. What might work is adding a '+' on the front, e.g.: mov +(4),r1 David
Hi David, 2017-04-18 18:57 GMT+09:00 David Howells <dhowells@redhat.com>: > Masahiro Yamada <yamada.masahiro@socionext.com> wrote: > >> This part ended up in redundant code after touched by multiple >> people. >> >> [1] Commit 3234282f33b2 ("x86, asm: Fix CFI macro invocations to >> deal with shortcomings in gas") added parentheses for defined >> expressions to support old gas for x86. >> >> [2] Commit a22dcdb0032c ("x86, asm: Fix ancient-GAS workaround") >> split the pattern into two to avoid parentheses for non-numeric >> expressions. >> >> [3] Commit 95a2f6f72d37 ("Partially revert patch that encloses >> asm-offset.h numbers in brackets") removed parentheses from numeric >> expressions as well because parentheses in MN10300 assembly have a >> special meaning (pointer access). >> >> Apparently, there is a conflict between [1] and [3]. After all, >> [3] took precedence, and a long time has passed since then. > > There's a conflict between [1] and various assembly code formats. Some > formats define, say, > > mov 4,r1 > > to move the number 4 into register r1, and: > > mov (4),r1 > > to move the contents of the memory at address 4 into r1. Therefore, you > cannot simply wrap numeric operands in brackets. What might work is adding a > '+' on the front, e.g.: > > mov +(4),r1 > > David OK, thanks for this info. But, nobody has raised a flag about the reverted [1]. (perhaps, nobody cares about the old gas any more?) So, I think this patch will be OK. BTW, do you still maintain mn10300? (Before, I worked for Panasonic, i.e. former Matsushita) -- Best Regards Masahiro Yamada
Masahiro Yamada <yamada.masahiro@socionext.com> wrote:
> BTW, do you still maintain mn10300?
Kind of. I don't really have working hardware any more, though.
David
Hi David, 2017-04-18 22:15 GMT+09:00 David Howells <dhowells@redhat.com>: > Masahiro Yamada <yamada.masahiro@socionext.com> wrote: > >> BTW, do you still maintain mn10300? > > Kind of. I don't really have working hardware any more, though. > > David Thanks for maintaining this. -- Best Regards Masahiro Yamada
diff --git a/scripts/Makefile.lib b/scripts/Makefile.lib index e36427a..9c20690 100644 --- a/scripts/Makefile.lib +++ b/scripts/Makefile.lib @@ -415,7 +415,6 @@ cmd_xzmisc = (cat $(filter-out FORCE,$^) | \ # Default sed regexp - multiline due to syntax constraints define sed-offsets "/^->/{s:->#\(.*\):/* \1 */:; \ - s:^->\([^ ]*\) [\$$#]*\([-0-9]*\) \(.*\):#define \1 \2 /* \3 */:; \ s:^->\([^ ]*\) [\$$#]*\([^ ]*\) \(.*\):#define \1 \2 /* \3 */:; \ s:->::; p;}" endef
This part ended up in redundant code after touched by multiple people. [1] Commit 3234282f33b2 ("x86, asm: Fix CFI macro invocations to deal with shortcomings in gas") added parentheses for defined expressions to support old gas for x86. [2] Commit a22dcdb0032c ("x86, asm: Fix ancient-GAS workaround") split the pattern into two to avoid parentheses for non-numeric expressions. [3] Commit 95a2f6f72d37 ("Partially revert patch that encloses asm-offset.h numbers in brackets") removed parentheses from numeric expressions as well because parentheses in MN10300 assembly have a special meaning (pointer access). Apparently, there is a conflict between [1] and [3]. After all, [3] took precedence, and a long time has passed since then. Now, merge the two patterns again because the first one is covered by the other. Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com> --- scripts/Makefile.lib | 1 - 1 file changed, 1 deletion(-) -- 2.7.4