[PATCH 2/2] x86_64: Use __seg_fs qualifiers in NPTL accessors
H.J. Lu
hjl.tools@gmail.com
Wed Aug 20 03:18:55 GMT 2025
On Tue, Aug 19, 2025 at 8:06 PM H.J. Lu <hjl.tools@gmail.com> wrote:
>
> On Sun, Aug 17, 2025 at 8:50 AM Uros Bizjak <ubizjak@gmail.com> wrote:
> >
> > Use __seg_fs named address space qualifiers to cast NPTL accessors
> > to %fs: prefixed addresses. Use volatile access only where
> > strictly necessary.
> >
> > Signed-off-by: Uros Bizjak <ubizjak@gmail.com>
> > Cc: H.J.Lu <hjl.tools@gmail.com>
> > Cc: Florian Weimer <fweimer@redhat.com>
> > Cc: Carlos O'Donell <carlos@redhat.com>
> > ---
> > sysdeps/x86_64/nptl/rseq-access.h | 82 ++++++++-----------
> > sysdeps/x86_64/nptl/tcb-access.h | 127 ++++++++----------------------
> > 2 files changed, 66 insertions(+), 143 deletions(-)
> >
> > diff --git a/sysdeps/x86_64/nptl/rseq-access.h b/sysdeps/x86_64/nptl/rseq-access.h
> > index bc966b2972..214cb96797 100644
> > --- a/sysdeps/x86_64/nptl/rseq-access.h
> > +++ b/sysdeps/x86_64/nptl/rseq-access.h
> > @@ -16,34 +16,38 @@
> > License along with the GNU C Library; if not, see
> > <https://www.gnu.org/licenses/>. */
> >
> > +/* Read member of the RSEQ area directly. */
> > +#define RSEQ_GETMEM(member) \
> > + ({ \
> > + _Static_assert (sizeof (RSEQ_SELF()->member) == 1 \
> > + || sizeof (RSEQ_SELF()->member) == 4 \
> > + || sizeof (RSEQ_SELF()->member) == 8, \
> > + "size of rseq data"); \
> > + (*(__typeof (RSEQ_SELF()->member) __seg_fs *) \
> > + (__rseq_offset + offsetof (struct rseq_area, member))); \
info: __rseq_offset: -192
%fs + negative offset doesn't work for x32.
> > + })
> > +
> > /* Read member of the RSEQ area directly, with single-copy atomicity semantics. */
> > #define RSEQ_GETMEM_ONCE(member) \
> > - ({ __typeof (RSEQ_SELF()->member) __value; \
> > - _Static_assert (sizeof (__value) == 1 \
> > - || sizeof (__value) == 4 \
> > - || sizeof (__value) == 8, \
> > - "size of rseq data"); \
> > - if (sizeof (__value) == 1) \
> > - asm volatile ("movb %%fs:%P2(%q3),%b0" \
> > - : "=q" (__value) \
> > - : "0" (0), "i" (offsetof (struct rseq_area, member)), \
> > - "r" ((long long int) __rseq_offset)); \
> > - else if (sizeof (__value) == 4) \
> > - asm volatile ("movl %%fs:%P1(%q2),%0" \
> > - : "=r" (__value) \
> > - : "i" (offsetof (struct rseq_area, member)), \
> > - "r" ((long long int) __rseq_offset)); \
> > - else /* 8 */ \
> > - { \
> > - asm volatile ("movq %%fs:%P1(%q2),%q0" \
> > - : "=r" (__value) \
> > - : "i" (offsetof (struct rseq_area, member)), \
> > - "r" ((long long int) __rseq_offset)); \
> > - } \
> > - __value; })
> > + ({ \
> > + _Static_assert (sizeof (RSEQ_SELF()->member) == 1 \
> > + || sizeof (RSEQ_SELF()->member) == 4 \
> > + || sizeof (RSEQ_SELF()->member) == 8, \
> > + "size of rseq data"); \
> > + (*(volatile __typeof (RSEQ_SELF()->member) __seg_fs *) \
> > + (__rseq_offset + offsetof (struct rseq_area, member))); \
> > + })
> >
> > -/* Read member of the RSEQ area directly. */
> > -#define RSEQ_GETMEM(member) RSEQ_GETMEM_ONCE(member)
> > +/* Set member of the RSEQ area directly. */
> > +#define RSEQ_SETMEM(member, value) \
> > + ({ \
> > + _Static_assert (sizeof (RSEQ_SELF()->member) == 1 \
> > + || sizeof (RSEQ_SELF()->member) == 4 \
> > + || sizeof (RSEQ_SELF()->member) == 8, \
> > + "size of rseq data"); \
> > + (*(__typeof (RSEQ_SELF()->member) __seg_fs *) \
> > + (__rseq_offset + offsetof (struct rseq_area, member)) = (value)); \
> > + })
> >
> > /* Set member of the RSEQ area directly, with single-copy atomicity semantics. */
> > #define RSEQ_SETMEM_ONCE(member, value) \
> > @@ -51,27 +55,7 @@
> > _Static_assert (sizeof (RSEQ_SELF()->member) == 1 \
> > || sizeof (RSEQ_SELF()->member) == 4 \
> > || sizeof (RSEQ_SELF()->member) == 8, \
> > - "size of rseq data"); \
> > - if (sizeof (RSEQ_SELF()->member) == 1) \
> > - asm volatile ("movb %b0,%%fs:%P1(%q2)" : \
> > - : "iq" (value), \
> > - "i" (offsetof (struct rseq_area, member)), \
> > - "r" ((long long int) __rseq_offset)); \
> > - else if (sizeof (RSEQ_SELF()->member) == 4) \
> > - asm volatile ("movl %0,%%fs:%P1(%q2)" : \
> > - : IMM_MODE (value), \
> > - "i" (offsetof (struct rseq_area, member)), \
> > - "r" ((long long int) __rseq_offset)); \
> > - else /* 8 */ \
> > - { \
> > - /* Since movq takes a signed 32-bit immediate or a register source \
> > - operand, use "er" constraint for 32-bit signed integer constant \
> > - or register. */ \
> > - asm volatile ("movq %q0,%%fs:%P1(%q2)" : \
> > - : "er" ((uint64_t) cast_to_integer (value)), \
> > - "i" (offsetof (struct rseq_area, member)), \
> > - "r" ((long long int) __rseq_offset)); \
> > - }})
> > -
> > -/* Set member of the RSEQ area directly. */
> > -#define RSEQ_SETMEM(member, value) RSEQ_SETMEM_ONCE(member, value)
> > + "size of rseq data"); \
> > + (*(volatile __typeof (RSEQ_SELF()->member) __seg_fs *) \
> > + (__rseq_offset + offsetof (struct rseq_area, member)) = (value)); \
> > + })
> > diff --git a/sysdeps/x86_64/nptl/tcb-access.h b/sysdeps/x86_64/nptl/tcb-access.h
> > index defd76f54f..445f328ac2 100644
> > --- a/sysdeps/x86_64/nptl/tcb-access.h
> > +++ b/sysdeps/x86_64/nptl/tcb-access.h
> > @@ -17,116 +17,55 @@
> > <https://www.gnu.org/licenses/>. */
> >
> > /* Read member of the thread descriptor directly. */
> > -# define THREAD_GETMEM(descr, member) \
> > - ({ __typeof (descr->member) __value; \
> > - _Static_assert (sizeof (__value) == 1 \
> > - || sizeof (__value) == 4 \
> > - || sizeof (__value) == 8, \
> > +#define THREAD_GETMEM(descr, member) \
> > + ({ \
> > + _Static_assert (sizeof (descr->member) == 1 \
> > + || sizeof (descr->member) == 4 \
> > + || sizeof (descr->member) == 8, \
> > "size of per-thread data"); \
> > - if (sizeof (__value) == 1) \
> > - asm volatile ("movb %%fs:%P2,%b0" \
> > - : "=q" (__value) \
> > - : "0" (0), "i" (offsetof (struct pthread, member))); \
> > - else if (sizeof (__value) == 4) \
> > - asm volatile ("movl %%fs:%P1,%0" \
> > - : "=r" (__value) \
> > - : "i" (offsetof (struct pthread, member))); \
> > - else /* 8 */ \
> > - { \
> > - asm volatile ("movq %%fs:%P1,%q0" \
> > - : "=r" (__value) \
> > - : "i" (offsetof (struct pthread, member))); \
> > - } \
> > - __value; })
> > + (*(__typeof (descr->member) __seg_fs *) \
> > + offsetof (struct pthread, member)); \
> > + })
> >
> > -/* THREAD_GETMEM already forces a read. */
> > -#define THREAD_GETMEM_VOLATILE(descr, member) THREAD_GETMEM (descr, member)
> > +#define THREAD_GETMEM_VOLATILE(descr, member) \
> > + ({ \
> > + _Static_assert (sizeof (descr->member) == 1 \
> > + || sizeof (descr->member) == 4 \
> > + || sizeof (descr->member) == 8, \
> > + "size of per-thread data"); \
> > + (*(volatile __typeof (descr->member) __seg_fs *) \
> > + offsetof (struct pthread, member)); \
> > + })
> >
> > /* Same as THREAD_GETMEM, but the member offset can be non-constant. */
> > -# define THREAD_GETMEM_NC(descr, member, idx) \
> > - ({ __typeof (descr->member[0]) __value; \
> > - _Static_assert (sizeof (__value) == 1 \
> > - || sizeof (__value) == 4 \
> > - || sizeof (__value) == 8, \
> > +#define THREAD_GETMEM_NC(descr, member, idx) \
> > + ({ \
> > + _Static_assert (sizeof (descr->member[0]) == 1 \
> > + || sizeof (descr->member[0]) == 4 \
> > + || sizeof (descr->member[0]) == 8, \
> > "size of per-thread data"); \
> > - if (sizeof (__value) == 1) \
> > - asm volatile ("movb %%fs:%P2(%q3),%b0" \
> > - : "=q" (__value) \
> > - : "0" (0), "i" (offsetof (struct pthread, member[0])), \
> > - "r" (idx)); \
> > - else if (sizeof (__value) == 4) \
> > - asm volatile ("movl %%fs:%P1(,%q2,4),%0" \
> > - : "=r" (__value) \
> > - : "i" (offsetof (struct pthread, member[0])), "r" (idx));\
> > - else /* 8 */ \
> > - { \
> > - asm volatile ("movq %%fs:%P1(,%q2,8),%q0" \
> > - : "=r" (__value) \
> > - : "i" (offsetof (struct pthread, member[0])), \
> > - "r" (idx)); \
> > - } \
> > - __value; })
> > -
> > -
> > -/* Loading addresses of objects on x86-64 needs to be treated special
> > - when generating PIC code. */
> > -#ifdef __pic__
> > -# define IMM_MODE "nr"
> > -#else
> > -# define IMM_MODE "ir"
> > -#endif
> > -
> > + (*(__typeof (descr->member[0]) __seg_fs *) \
> > + offsetof (struct pthread, member[idx])); \
> > + })
> >
> > /* Set member of the thread descriptor directly. */
> > -# define THREAD_SETMEM(descr, member, value) \
> > +#define THREAD_SETMEM(descr, member, value) \
> > ({ \
> > _Static_assert (sizeof (descr->member) == 1 \
> > || sizeof (descr->member) == 4 \
> > || sizeof (descr->member) == 8, \
> > "size of per-thread data"); \
> > - if (sizeof (descr->member) == 1) \
> > - asm volatile ("movb %b0,%%fs:%P1" : \
> > - : "iq" (value), \
> > - "i" (offsetof (struct pthread, member))); \
> > - else if (sizeof (descr->member) == 4) \
> > - asm volatile ("movl %0,%%fs:%P1" : \
> > - : IMM_MODE (value), \
> > - "i" (offsetof (struct pthread, member))); \
> > - else /* 8 */ \
> > - { \
> > - /* Since movq takes a signed 32-bit immediate or a register source \
> > - operand, use "er" constraint for 32-bit signed integer constant \
> > - or register. */ \
> > - asm volatile ("movq %q0,%%fs:%P1" : \
> > - : "er" ((uint64_t) cast_to_integer (value)), \
> > - "i" (offsetof (struct pthread, member))); \
> > - }})
> > -
> > + (*(__typeof (descr->member) __seg_fs *) \
> > + offsetof (struct pthread, member) = (value)); \
> > + })
> >
> > /* Same as THREAD_SETMEM, but the member offset can be non-constant. */
> > -# define THREAD_SETMEM_NC(descr, member, idx, value) \
> > +#define THREAD_SETMEM_NC(descr, member, idx, value) \
> > ({ \
> > _Static_assert (sizeof (descr->member[0]) == 1 \
> > || sizeof (descr->member[0]) == 4 \
> > || sizeof (descr->member[0]) == 8, \
> > "size of per-thread data"); \
> > - if (sizeof (descr->member[0]) == 1) \
> > - asm volatile ("movb %b0,%%fs:%P1(%q2)" : \
> > - : "iq" (value), \
> > - "i" (offsetof (struct pthread, member[0])), \
> > - "r" (idx)); \
> > - else if (sizeof (descr->member[0]) == 4) \
> > - asm volatile ("movl %0,%%fs:%P1(,%q2,4)" : \
> > - : IMM_MODE (value), \
> > - "i" (offsetof (struct pthread, member[0])), \
> > - "r" (idx)); \
> > - else /* 8 */ \
> > - { \
> > - /* Since movq takes a signed 32-bit immediate or a register source \
> > - operand, use "er" constraint for 32-bit signed integer constant \
> > - or register. */ \
> > - asm volatile ("movq %q0,%%fs:%P1(,%q2,8)" : \
> > - : "er" ((uint64_t) cast_to_integer (value)), \
> > - "i" (offsetof (struct pthread, member[0])), \
> > - "r" (idx)); \
> > - }})
> > + (*(__typeof (descr->member[0]) __seg_fs *) \
> > + offsetof (struct pthread, member[idx]) = (value)); \
> > + })
> > --
> > 2.50.1
> >
>
> This doesn't work for x32:
>
> ../localedata/gen-locale.sh: line 29: 3948007 Segmentation fault
> (core dumped) ${localedef_before_env} ${run_program_env}
> I18NPATH=../localedata ${localedef_after_env} $flags -f $charmap -i
> $input ${common_objpfx}localedata/$out
> Charmap: "UTF-8" Inputfile: "C" Outputdir: "C.UTF-8" failed
> /bin/sh: line 4:
> /export/build/gnu/tools-build/glibc-x32-cet-gitlab/build-x86_64-linux/localedata/C.UTF-8/LC_CTYPE.test-result:
> No such file or directory
> make[4]: *** [../gen-locales.mk:28:
> /export/build/gnu/tools-build/glibc-x32-cet-gitlab/build-x86_64-linux/localedata/C.UTF-8/LC_CTYPE]
> Error 1
>
>
> --
> H.J.
--
H.J.
More information about the Libc-alpha
mailing list