[PATCH v4 0/3] Restore the original macro behaviors
H.J. Lu
hjl.tools@gmail.com
Mon Aug 12 16:17:38 GMT 2024
On Mon, Aug 12, 2024 at 8:49 AM Jan Beulich <jbeulich@suse.com> wrote:
>
> On 12.08.2024 17:43, H.J. Lu wrote:
> > Changes in v4:
> >
> > 1. Add inner white spaces to arg1 test comments.
> >
> > Changes in v3:
> >
> > 1. Add test for a white space before argument added by C preprocessor to
> > macros/arg1.s.
> > 2. Add comments to tests.
> >
> > Changes in v2:
> >
> > 1. Mention PR gas/32073 in reverting commits.
> > 2. Add a test from x86-64 glibc.
> >
> > ----
> > commit 6ae8a30d44f016cafb46a75843b5109316eb1996
> > Author: Jan Beulich <jbeulich@suse.com>
> > Date: Fri Aug 9 11:59:31 2024 +0200
> >
> > gas: have scrubber retain more whitespace
> >
> > breaks GCC builds for ARM, AVR, PRU and others as well as x86-64 kernel
> > and glibc builds. Revert
> >
> > 7dd0dfbde7e gas: drop scrubber states 14 and 15
> > 6ae8a30d44f gas: have scrubber retain more whitespace
> >
> > to restore the original macro behaviors and add tests from Linux kernel
> > and glibc.
> >
> > H.J. Lu (3):
> > Revert "gas: drop scrubber states 14 and 15"
> > Revert "gas: have scrubber retain more whitespace"
> > gas: Add macro tests for PR gas/32073
> >
> > gas/NEWS | 7 -
> > gas/app.c | 200 ++++++++++++++++++------
> > gas/config/tc-aarch64.c | 19 +--
> > gas/config/tc-arm.c | 2 -
> > gas/config/tc-crx.c | 6 +-
> > gas/config/tc-csky.c | 23 +--
> > gas/config/tc-ia64.h | 3 +
> > gas/config/tc-pru.c | 10 +-
> > gas/config/tc-sparc.c | 48 ++----
> > gas/config/tc-tic6x.h | 2 +
> > gas/config/tc-v850.c | 4 -
> > gas/testsuite/gas/all/macro.l | 10 --
> > gas/testsuite/gas/all/macro.s | 4 +-
> > gas/testsuite/gas/i386/x86-64-apx-nf.s | 40 ++---
> > gas/testsuite/gas/i386/x86-64-macro-1.d | 11 ++
> > gas/testsuite/gas/i386/x86-64-macro-1.s | 9 ++
> > gas/testsuite/gas/i386/x86-64.exp | 2 +
> > gas/testsuite/gas/macros/arg1.d | 7 +
> > gas/testsuite/gas/macros/arg1.s | 12 ++
> > gas/testsuite/gas/macros/macros.exp | 2 +
> > opcodes/cgen-asm.in | 52 +++---
> > opcodes/epiphany-asm.c | 52 +++---
> > opcodes/fr30-asm.c | 52 +++---
> > opcodes/frv-asm.c | 52 +++---
> > opcodes/ip2k-asm.c | 52 +++---
> > opcodes/iq2000-asm.c | 52 +++---
> > opcodes/lm32-asm.c | 52 +++---
> > opcodes/m32c-asm.c | 52 +++---
> > opcodes/m32r-asm.c | 52 +++---
> > opcodes/mep-asm.c | 52 +++---
> > opcodes/mt-asm.c | 52 +++---
> > opcodes/nds32-asm.c | 5 -
> > opcodes/or1k-asm.c | 52 +++---
> > opcodes/xstormy16-asm.c | 52 +++---
> > 34 files changed, 482 insertions(+), 620 deletions(-)
> > create mode 100644 gas/testsuite/gas/i386/x86-64-macro-1.d
> > create mode 100644 gas/testsuite/gas/i386/x86-64-macro-1.s
> > create mode 100644 gas/testsuite/gas/macros/arg1.d
> > create mode 100644 gas/testsuite/gas/macros/arg1.s
>
> I'm willing to give my okay here as long as the discussion on what to do
> longer term won't die out. With this reverted I would work on a fix for
The sensible solution is to document these longstanding usages
as features with tests.
> the one case that I think we really need to deal with (blank between
> backslash and parameter name). What else (if anything) to change before
> making a 2nd attempt depends on outcomes of the discussion (and perhaps
> Nick's input). Leaving things as broken as they are is not an option to
> me.
>
> Jan
--
H.J.
More information about the Binutils
mailing list