| Message ID | 1407556266-12337-2-git-send-email-apinski@cavium.com |
|---|---|
| State | New |
| Headers | show |
On Fri, Aug 8, 2014 at 8:51 PM, Andrew Pinski <apinski@cavium.com> wrote: > OK? When the second patch is approved? Ping? > > Thanks, > Andrew Pinski > > ChangeLog: > Revert: > 2011-08-19 H.J. Lu <hongjiu.lu@intel.com> > > PR middle-end/49721 > * explow.c (convert_memory_address_addr_space): Also permute the > conversion and addition of constant for zero-extend. > > --- > gcc/explow.c | 19 +++++++------------ > 1 files changed, 7 insertions(+), 12 deletions(-) > > diff --git a/gcc/explow.c b/gcc/explow.c > index 92c4e57..eb7dc85 100644 > --- a/gcc/explow.c > +++ b/gcc/explow.c > @@ -376,23 +376,18 @@ convert_memory_address_addr_space (enum machine_mode to_mode ATTRIBUTE_UNUSED, > > case PLUS: > case MULT: > - /* FIXME: For addition, we used to permute the conversion and > - addition operation only if one operand is a constant and > - converting the constant does not change it or if one operand > - is a constant and we are using a ptr_extend instruction > - (POINTERS_EXTEND_UNSIGNED < 0) even if the resulting address > - may overflow/underflow. We relax the condition to include > - zero-extend (POINTERS_EXTEND_UNSIGNED > 0) since the other > - parts of the compiler depend on it. See PR 49721. > - > + /* For addition we can safely permute the conversion and addition > + operation if one operand is a constant and converting the constant > + does not change it or if one operand is a constant and we are > + using a ptr_extend instruction (POINTERS_EXTEND_UNSIGNED < 0). > We can always safely permute them if we are making the address > narrower. */ > if (GET_MODE_SIZE (to_mode) < GET_MODE_SIZE (from_mode) > || (GET_CODE (x) == PLUS > && CONST_INT_P (XEXP (x, 1)) > - && (POINTERS_EXTEND_UNSIGNED != 0 > - || XEXP (x, 1) == convert_memory_address_addr_space > - (to_mode, XEXP (x, 1), as)))) > + && (XEXP (x, 1) == convert_memory_address_addr_space > + (to_mode, XEXP (x, 1), as) > + || POINTERS_EXTEND_UNSIGNED < 0))) > return gen_rtx_fmt_ee (GET_CODE (x), to_mode, > convert_memory_address_addr_space > (to_mode, XEXP (x, 0), as), > -- > 1.7.2.5 >
On Tue, Oct 14, 2014 at 12:35 AM, Andrew Pinski <pinskia@gmail.com> wrote: > On Fri, Aug 8, 2014 at 8:51 PM, Andrew Pinski <apinski@cavium.com> wrote: >> OK? When the second patch is approved? > > Ping? Ok if the second patch was approved. Richard. >> >> Thanks, >> Andrew Pinski >> >> ChangeLog: >> Revert: >> 2011-08-19 H.J. Lu <hongjiu.lu@intel.com> >> >> PR middle-end/49721 >> * explow.c (convert_memory_address_addr_space): Also permute the >> conversion and addition of constant for zero-extend. >> >> --- >> gcc/explow.c | 19 +++++++------------ >> 1 files changed, 7 insertions(+), 12 deletions(-) >> >> diff --git a/gcc/explow.c b/gcc/explow.c >> index 92c4e57..eb7dc85 100644 >> --- a/gcc/explow.c >> +++ b/gcc/explow.c >> @@ -376,23 +376,18 @@ convert_memory_address_addr_space (enum machine_mode to_mode ATTRIBUTE_UNUSED, >> >> case PLUS: >> case MULT: >> - /* FIXME: For addition, we used to permute the conversion and >> - addition operation only if one operand is a constant and >> - converting the constant does not change it or if one operand >> - is a constant and we are using a ptr_extend instruction >> - (POINTERS_EXTEND_UNSIGNED < 0) even if the resulting address >> - may overflow/underflow. We relax the condition to include >> - zero-extend (POINTERS_EXTEND_UNSIGNED > 0) since the other >> - parts of the compiler depend on it. See PR 49721. >> - >> + /* For addition we can safely permute the conversion and addition >> + operation if one operand is a constant and converting the constant >> + does not change it or if one operand is a constant and we are >> + using a ptr_extend instruction (POINTERS_EXTEND_UNSIGNED < 0). >> We can always safely permute them if we are making the address >> narrower. */ >> if (GET_MODE_SIZE (to_mode) < GET_MODE_SIZE (from_mode) >> || (GET_CODE (x) == PLUS >> && CONST_INT_P (XEXP (x, 1)) >> - && (POINTERS_EXTEND_UNSIGNED != 0 >> - || XEXP (x, 1) == convert_memory_address_addr_space >> - (to_mode, XEXP (x, 1), as)))) >> + && (XEXP (x, 1) == convert_memory_address_addr_space >> + (to_mode, XEXP (x, 1), as) >> + || POINTERS_EXTEND_UNSIGNED < 0))) >> return gen_rtx_fmt_ee (GET_CODE (x), to_mode, >> convert_memory_address_addr_space >> (to_mode, XEXP (x, 0), as), >> -- >> 1.7.2.5 >>
On 10/13/14 16:35, Andrew Pinski wrote: > On Fri, Aug 8, 2014 at 8:51 PM, Andrew Pinski <apinski@cavium.com> wrote: >> OK? When the second patch is approved? > > Ping? > >> >> Thanks, >> Andrew Pinski >> >> ChangeLog: >> Revert: >> 2011-08-19 H.J. Lu <hongjiu.lu@intel.com> >> >> PR middle-end/49721 >> * explow.c (convert_memory_address_addr_space): Also permute the >> conversion and addition of constant for zero-extend. OK if/when 2nd patch is approved. As far as I can tell, it hasn't been approved yet. jeff
On Wed, Oct 15, 2014 at 10:29 AM, Jeff Law <law@redhat.com> wrote: > On 10/13/14 16:35, Andrew Pinski wrote: >> >> On Fri, Aug 8, 2014 at 8:51 PM, Andrew Pinski <apinski@cavium.com> wrote: >>> >>> OK? When the second patch is approved? >> >> >> Ping? >> >>> >>> Thanks, >>> Andrew Pinski >>> >>> ChangeLog: >>> Revert: >>> 2011-08-19 H.J. Lu <hongjiu.lu@intel.com> >>> >>> PR middle-end/49721 >>> * explow.c (convert_memory_address_addr_space): Also permute the >>> conversion and addition of constant for zero-extend. > > OK if/when 2nd patch is approved. As far as I can tell, it hasn't been > approved yet. Both patches were approved yesterday. Richard Biener approved this one and Richard Henderson approved the second patch with a minor change. Thanks, Andrew > > jeff >
On 10/13/14 16:35, Andrew Pinski wrote: > On Fri, Aug 8, 2014 at 8:51 PM, Andrew Pinski <apinski@cavium.com> wrote: >> OK? When the second patch is approved? > > Ping? Nevermind my comments, both were approved in the last day or two. jeff
diff --git a/gcc/explow.c b/gcc/explow.c index 92c4e57..eb7dc85 100644 --- a/gcc/explow.c +++ b/gcc/explow.c @@ -376,23 +376,18 @@ convert_memory_address_addr_space (enum machine_mode to_mode ATTRIBUTE_UNUSED, case PLUS: case MULT: - /* FIXME: For addition, we used to permute the conversion and - addition operation only if one operand is a constant and - converting the constant does not change it or if one operand - is a constant and we are using a ptr_extend instruction - (POINTERS_EXTEND_UNSIGNED < 0) even if the resulting address - may overflow/underflow. We relax the condition to include - zero-extend (POINTERS_EXTEND_UNSIGNED > 0) since the other - parts of the compiler depend on it. See PR 49721. - + /* For addition we can safely permute the conversion and addition + operation if one operand is a constant and converting the constant + does not change it or if one operand is a constant and we are + using a ptr_extend instruction (POINTERS_EXTEND_UNSIGNED < 0). We can always safely permute them if we are making the address narrower. */ if (GET_MODE_SIZE (to_mode) < GET_MODE_SIZE (from_mode) || (GET_CODE (x) == PLUS && CONST_INT_P (XEXP (x, 1)) - && (POINTERS_EXTEND_UNSIGNED != 0 - || XEXP (x, 1) == convert_memory_address_addr_space - (to_mode, XEXP (x, 1), as)))) + && (XEXP (x, 1) == convert_memory_address_addr_space + (to_mode, XEXP (x, 1), as) + || POINTERS_EXTEND_UNSIGNED < 0))) return gen_rtx_fmt_ee (GET_CODE (x), to_mode, convert_memory_address_addr_space (to_mode, XEXP (x, 0), as),
OK? When the second patch is approved? Thanks, Andrew Pinski ChangeLog: Revert: 2011-08-19 H.J. Lu <hongjiu.lu@intel.com> PR middle-end/49721 * explow.c (convert_memory_address_addr_space): Also permute the conversion and addition of constant for zero-extend. --- gcc/explow.c | 19 +++++++------------ 1 files changed, 7 insertions(+), 12 deletions(-)