https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64243
--- Comment #5 from Uroš Bizjak ---
This is fixed in gcc-11:
--cut here--
struct TestFloat { float x; };
struct TestDouble { double x; };
struct TestFloat foo (struct TestFloat x) { return x; }
struct TestDouble bar (struct TestDouble x) {
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64243
Uroš Bizjak changed:
What|Removed |Added
CC||10walls at gmail dot com
--- Comment #4
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98439
--- Comment #3 from Uroš Bizjak ---
I don't think this is a backend bug. The position of split pass in the pass
sequence assumes that no split candidates will be emitted after regstack, as
can be seen from the gate function of the
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97684
Uroš Bizjak changed:
What|Removed |Added
Component|target |rtl-optimization
Ever confirmed|0
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98439
Uroš Bizjak changed:
What|Removed |Added
Component|target |rtl-optimization
Last reconfirmed|
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92658
--- Comment #22 from Uroš Bizjak ---
(In reply to Hongtao.liu from comment #21)
> Add define_code_attr like aarch64/iterators.md?
>
> --
> ;; Map rtl objects to optab names
> (define_code_attr optab [(ashift "ashl")
>
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96793
Uroš Bizjak changed:
What|Removed |Added
Resolution|--- |FIXED
Status|ASSIGNED
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96793
--- Comment #22 from Uroš Bizjak ---
https://gcc.gnu.org/g:edb28850520d1137d12a1cc1c0e89c11e6b0c6ef
commit r8-10691-gedb28850520d1137d12a1cc1c0e89c11e6b0c6ef
Author: Uros Bizjak
Date: Wed Dec 23 09:18:12 2020 +0100
i386: Fix
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96793
--- Comment #21 from Uroš Bizjak ---
https://gcc.gnu.org/g:c40b640ebcef1aae78eaca56e04d204dda9e4cad
commit r9-9126-gc40b640ebcef1aae78eaca56e04d204dda9e4cad
Author: Uros Bizjak
Date: Wed Dec 23 09:09:29 2020 +0100
i386: Fix
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96793
--- Comment #20 from Uroš Bizjak ---
https://gcc.gnu.org/g:0bf0e0b86d3e2f12555479096baaf0ca7a9f7ac6
commit r10-9164-g0bf0e0b86d3e2f12555479096baaf0ca7a9f7ac6
Author: Uros Bizjak
Date: Tue Dec 22 21:11:51 2020 +0100
i386: Fix
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96793
--- Comment #19 from Uroš Bizjak ---
https://gcc.gnu.org/g:337ed0eb490b14899f4049bc4c8922eb1d8a2e67
commit r11-6303-g337ed0eb490b14899f4049bc4c8922eb1d8a2e67
Author: Uros Bizjak
Date: Tue Dec 22 18:13:24 2020 +0100
i386: Fix
Component: middle-end
Assignee: unassigned at gcc dot gnu.org
Reporter: ubizjak at gmail dot com
Target Milestone: ---
Split out from PR96793, where Marc says in c13:
-q-
x-x does depend on the rounding mode (the transformation in match.pd gets it
wrong, by the way).
-/q
at gcc dot gnu.org |ubizjak at gmail dot com
--- Comment #18 from Uroš Bizjak ---
Created attachment 49833
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=49833=edit
Proposed patch
Proposed patch that removes the sign from a temporary
with FE_DOWNWARD, where x - x = -0.0
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98060
Uroš Bizjak changed:
What|Removed |Added
Target Milestone|--- |12.0
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98375
Uroš Bizjak changed:
What|Removed |Added
Status|UNCONFIRMED |NEW
Last reconfirmed|
Assignee: unassigned at gcc dot gnu.org
Reporter: ubizjak at gmail dot com
Target Milestone: ---
This metabug is used to track all the patches which have been written during
Stage 3 of GCC 11 but do not qualify for that stage, and are waiting for Stage
1 of GCC 12 to be applied
-*-* i?86-*-* |x86
Status|NEW |ASSIGNED
Host|x86_64-pc-linux-gnu |
Target Milestone|--- |12.0
Assignee|unassigned at gcc dot gnu.org |ubizjak at gmail dot com
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98218
--- Comment #3 from Uroš Bizjak ---
Testcase 1:
--cut here--
typedef short vec __attribute__((vector_size(8)));
typedef unsigned short uvec __attribute__((vector_size(8)));
vec lt (vec a, vec b) { return a < b; }
vec le (vec a, vec b) { return
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98218
--- Comment #2 from Uroš Bizjak ---
Created attachment 49796
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=49796=edit
Proposed patch to implement integer vector compares
Attached patch implements integer vector compares.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98169
--- Comment #7 from Uroš Bizjak ---
(In reply to Jakub Jelinek from comment #6)
> Not familiar with the 64-bit vector support myself, CCing Uros on that.
PR98218
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=91384
--- Comment #8 from Uroš Bizjak ---
(In reply to Jakub Jelinek from comment #1)
> Started with r223689. Though, generally that change looks like a useful
> GIMPLE canonicalization.
How about we amend the above change to:
diff --git
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=91384
--- Comment #7 from Uroš Bizjak ---
Still happens on trunk.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=78952
Uroš Bizjak changed:
What|Removed |Added
Status|NEW |RESOLVED
Resolution|---
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=78952
Uroš Bizjak changed:
What|Removed |Added
Target Milestone|--- |9.0
--- Comment #7 from Uroš Bizjak ---
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95046
Uroš Bizjak changed:
What|Removed |Added
Status|ASSIGNED|RESOLVED
Resolution|---
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95750
Uroš Bizjak changed:
What|Removed |Added
Target Milestone|--- |11.0
Resolution|---
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98212
--- Comment #2 from Uroš Bizjak ---
f1 is currently unoptimal by design, the compiler is unable to merge trapping
and non-trapping instructions. There is already a PR for that.
f2 is not optimal. The conditional jump to the unconditional jump
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92469
--- Comment #11 from Uroš Bizjak ---
*** Bug 98194 has been marked as a duplicate of this bug. ***
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98194
Uroš Bizjak changed:
What|Removed |Added
Resolution|--- |DUPLICATE
Status|UNCONFIRMED
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98178
--- Comment #2 from Uroš Bizjak ---
On a related note, the combine splitter is a very mysterious beast, and does
not easily tell, why the particular combination is rejected. Without any debug
in debug logs it is very frustrating to figure out
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98178
--- Comment #1 from Uroš Bizjak ---
The attached patch with the following testcase:
--cut here--
int test (int a, int b)
{
return a << (b & 31);
}
--cut here--
fails to generate a single shift insn, because it does not trigger the call to
Component: rtl-optimization
Assignee: unassigned at gcc dot gnu.org
Reporter: ubizjak at gmail dot com
Target Milestone: ---
Created attachment 49697
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=49697=edit
Proof-of-concept target patch to reimplement shift insn splitt
|RESOLVED
Assignee|jakub at gcc dot gnu.org |ubizjak at gmail dot com
--- Comment #8 from Uroš Bizjak ---
Fixed for gcc-9.4, gcc-10.3 and gcc-11.0.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98086
--- Comment #7 from Uroš Bizjak ---
https://gcc.gnu.org/g:521c839fad4e4a30cdadda254fb3f07706285033
commit r9-9096-g521c839fad4e4a30cdadda254fb3f07706285033
Author: Uros Bizjak
Date: Thu Dec 3 19:08:23 2020 +0100
i386: Fix up
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98086
--- Comment #6 from Uroš Bizjak ---
https://gcc.gnu.org/g:4fa2ab664324eca33bd67981698c453820f70b86
commit r10-9116-g4fa2ab664324eca33bd67981698c453820f70b86
Author: Uros Bizjak
Date: Thu Dec 3 17:49:42 2020 +0100
i386: Fix up
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98086
--- Comment #5 from Uroš Bizjak ---
https://gcc.gnu.org/g:756f55e62f73eb32787497eb9e564d4b21a6e637
commit r11-5713-g756f55e62f73eb32787497eb9e564d4b21a6e637
Author: Uros Bizjak
Date: Thu Dec 3 17:49:42 2020 +0100
i386: Fix up
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98086
Uroš Bizjak changed:
What|Removed |Added
Attachment #49661|0 |1
is obsolete|
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98060
Uroš Bizjak changed:
What|Removed |Added
Attachment #49663|0 |1
is obsolete|
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98060
Uroš Bizjak changed:
What|Removed |Added
Assignee|unassigned at gcc dot gnu.org |ubizjak at gmail dot com
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98060
--- Comment #2 from Uroš Bizjak ---
Created attachment 49662
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=49662=edit
Testcases
: enhancement
Priority: P3
Component: tree-optimization
Assignee: unassigned at gcc dot gnu.org
Reporter: ubizjak at gmail dot com
Target Milestone: ---
These two functions are equivalent:
int t3 (float a, float b)
{
return __builtin_isless (a, b
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98079
Uroš Bizjak changed:
What|Removed |Added
Status|ASSIGNED|RESOLVED
Resolution|---
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98079
--- Comment #3 from Uroš Bizjak ---
https://gcc.gnu.org/g:018248ef6d03ca0088d5928928f966df99af134c
commit r11-5644-g018248ef6d03ca0088d5928928f966df99af134c
Author: Uros Bizjak
Date: Wed Dec 2 08:15:03 2020 +0100
i386: Fix abs an maxmin
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98036
Uroš Bizjak changed:
What|Removed |Added
Status|NEW |RESOLVED
Resolution|---
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98036
--- Comment #8 from Uroš Bizjak ---
Fixed by the above change for gcc-10 and gcc-11.
(I don't know why cvs-commit hook hates my commits...)
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98036
--- Comment #7 from Uroš Bizjak ---
https://gcc.gnu.org/g:67138ea1b041fe1637da6b3568f064f172d9358c
commit r11-5501-g67138ea1b041fe1637da6b3568f064f172d9358c
Author: Uros Bizjak
Date: Fri Nov 27 18:41:56 2020 +0100
testsuite/i386: Fix
at gcc dot gnu.org |ubizjak at gmail dot com
--- Comment #2 from Uroš Bizjak ---
Created attachment 49660
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=49660=edit
Proposed patch
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98079
--- Comment #1 from Uroš Bizjak ---
Have to disable expanders for TARGET_PARTIAL_REG_STALL.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98060
Uroš Bizjak changed:
What|Removed |Added
Severity|normal |enhancement
Last reconfirmed|
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98048
Uroš Bizjak changed:
What|Removed |Added
Component|target |tree-optimization
--- Comment #2 from
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98036
--- Comment #5 from Uroš Bizjak ---
Oh...
--cut here--
diff --git a/gcc/testsuite/gcc.target/i386/xop-hsubX.c
b/gcc/testsuite/gcc.target/i386/xop-hsubX.c
index f0fa9b312f2..dc7944d8bb7 100644
--- a/gcc/testsuite/gcc.target/i386/xop-hsubX.c
+++
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98036
--- Comment #4 from Uroš Bizjak ---
Something is wrong in [printf added by me]:
static int
check_sword2dword ()
{
int i, j, s, t, check_fails = 0;
for (i = 0; i < (NUM * 8); i = i + 8)
{
for (j = 0; j < 4; j++)
{
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98036
Uroš Bizjak changed:
What|Removed |Added
Last reconfirmed||2020-11-27
Status|UNCONFIRMED
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98036
--- Comment #2 from Uroš Bizjak ---
Created attachment 49637
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=49637=edit
Testcase with emulated XOP instructions
This testcase without XOP insns also fails with "-O2 -msse" on 32bit target.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98018
--- Comment #2 from Uroš Bizjak ---
I vote for -fforce-frame-pointer.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94846
Uroš Bizjak changed:
What|Removed |Added
Ever confirmed|0 |1
Last reconfirmed|
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97992
--- Comment #1 from Uroš Bizjak ---
This is expected with invalid asm.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97950
--- Comment #3 from Uroš Bizjak ---
Comment on attachment 49613
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=49613
gcc11-pr97950.patch
>+(define_insn_and_split "*setcc_hi_1"
>+ [(set (match_operand:HI 0 "register_operand" "=q")
>+
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96272
--- Comment #4 from Uroš Bizjak ---
(In reply to Jakub Jelinek from comment #2)
> Well, it needs the addition too, so I think this can't be done in match.pd,
> but would need to be done in some other pass (not sure which, perhaps
> phiopt?).
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96189
Uroš Bizjak changed:
What|Removed |Added
Status|REOPENED|RESOLVED
Resolution|---
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97873
--- Comment #11 from Uroš Bizjak ---
For the record, the removal of compare triggers:
- for linux x86_64 defconfig: 93 times
- for x86_64 GCC bootstrap: 360 times
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92651
--- Comment #11 from Uroš Bizjak ---
(In reply to rguent...@suse.de from comment #8)
> Sure. Another option would be to enhance STV even further
> (or add some peephole patterns - combine runs before STV2) to
> transform the
>
> psubd
dot gnu.org |ubizjak at gmail dot com
Resolution|--- |FIXED
Status|NEW |RESOLVED
--- Comment #10 from Uroš Bizjak ---
Fixed.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97873
--- Comment #9 from Uroš Bizjak ---
Fixed by:
commit fdace7584056de2f63bde2e3087f26beb6b0f97d
Author: Uros Bizjak
Date: Fri Nov 20 10:26:34 2020 +0100
i386: Optimize abs expansion [PR97873]
The patch introduces absM named pattern
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97873
Uroš Bizjak changed:
What|Removed |Added
Attachment #49588|0 |1
is obsolete|
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97873
--- Comment #7 from Uroš Bizjak ---
(In reply to Jakub Jelinek from comment #4)
> So then either we should expand the SWI48x mode abs for !TARGET_EXPAND_ABS
> into
> a pre-reload define_insn_and_split with abs that we'd split almost like
> smax,
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97873
--- Comment #6 from Uroš Bizjak ---
The attached patch generates:
movl%edi, %eax
negl%eax
cmovs %edi, %eax
ret
The patch changes CC mode of NEG instruction to CCGOCmode, which is the same
mode as the
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97873
--- Comment #5 from Uroš Bizjak ---
Created attachment 49588
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=49588=edit
Proposed patch
Attached patch introduces relevant peephole2 pattern (and fixes some other
issues).
at gcc dot gnu.org |ubizjak at gmail dot com
--- Comment #7 from Uroš Bizjak ---
I'll fix this.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97887
--- Comment #5 from Uroš Bizjak ---
> > This should have the following insn constraint:
> >
> > "TARGET_80387 && !(SSE_FLOAT_MODE_P (mode) && TARGET_SSE_MATH)"
> >
> > to hide it from combine in cases where relevant SSE mode is available.
>
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97887
--- Comment #3 from Uroš Bizjak ---
(In reply to Richard Biener from comment #2)
> combine first makes recog pick negsf2_i387_1:
This should have the following insn constraint:
"TARGET_80387 && !(SSE_FLOAT_MODE_P (mode) && TARGET_SSE_MATH)"
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=9
Uroš Bizjak changed:
What|Removed |Added
CC||qing.zhao at oracle dot com
Last
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97770
--- Comment #8 from Uroš Bizjak ---
(In reply to Richard Biener from comment #4)
> What's missing is middle-end folding support to narrow popcount to the
> appropriate internal function call with byte/half-word width when target
> support
> is
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97715
--- Comment #16 from Uroš Bizjak ---
(In reply to Jakub Jelinek from comment #15)
> They aren't live. But that loop checks that only if only_used is true, when
> one uses =all, it marks all regs that aren't fixed, aren't live at the end
> of
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97715
--- Comment #14 from Uroš Bizjak ---
(In reply to Jakub Jelinek from comment #11)
> I think you should do:
> --- gcc/function.c2020-10-31 17:41:19.756740009 +0100
> +++ gcc/function.c2020-11-04 17:02:51.199298173 +0100
> @@ -5871,6
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97715
--- Comment #12 from Uroš Bizjak ---
(In reply to Qing Zhao from comment #10)
> > On Nov 4, 2020, at 9:45 AM, ubizjak at gmail dot com
> > wrote:
> >> fixed registers should already be excluded from zeroing.
> >&g
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97715
--- Comment #9 from Uroš Bizjak ---
(In reply to qinzhao from comment #6)
> (In reply to Jakub Jelinek from comment #3)
> > ;; Floating-point register constraints.
> > (define_register_constraint "f"
> > "TARGET_80387 ||
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97715
--- Comment #7 from Uroš Bizjak ---
(In reply to qinzhao from comment #5)
> (In reply to H.J. Lu from comment #2)
> > (In reply to qinzhao from comment #1)
> > > for -fzero-call-used-regs=all, when zeroing st/mm registers under x87 exit
> > >
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97688
--- Comment #4 from Uroš Bizjak ---
(In reply to Richard Biener from comment #3)
> Ah. So I guess
>
> diff --git a/gcc/testsuite/gcc.dg/vect/tree-vect.h
> b/gcc/testsuite/gcc.dg/vect/tree-vect.h
> index 5d8d9eba3f8..c4b81441216 100644
> ---
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97688
--- Comment #1 from Uroš Bizjak ---
AVX2 should be detected using __get_cpuid_count, because a sub-leaf needs to be
specified for leaf 7.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97129
Uroš Bizjak changed:
What|Removed |Added
Resolution|--- |INVALID
Status|NEW
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97073
Uroš Bizjak changed:
What|Removed |Added
Component|target |middle-end
--- Comment #2 from Uroš
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96793
Uroš Bizjak changed:
What|Removed |Added
Assignee|ubizjak at gmail dot com |unassigned at gcc dot
gnu.org
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96793
--- Comment #12 from Uroš Bizjak ---
(In reply to Richard Biener from comment #10)
> (In reply to Uroš Bizjak from comment #7)
> > Created attachment 49144 [details]
> > Proposed patch
> >
> > Patch in testing.
>
> OTOH we _do_ try to
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96793
--- Comment #11 from Uroš Bizjak ---
Created attachment 49146
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=49146=edit
Testcase, suitable for gcc testsuite
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96793
--- Comment #9 from Uroš Bizjak ---
(In reply to Paweł Bylica from comment #8)
> Did you consider fixing the __builtin_floor() implementation?
No, because you can use -msse4 to generate ROUNDxx instructions.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96793
Uroš Bizjak changed:
What|Removed |Added
CC|uros at gcc dot gnu.org|
Target Milestone|---
at gcc dot gnu.org |ubizjak at gmail dot com
--- Comment #7 from Uroš Bizjak ---
Created attachment 49144
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=49144=edit
Proposed patch
Patch in testing.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96744
--- Comment #11 from Uroš Bizjak ---
(In reply to Hongtao.liu from comment #9)
> > > However, the patch assumes that avx512vp2intersect implies mavx512dq,
> >
> > Let me check this part.
>
> Confirmed, avx512vp2intersect implies avx512dq.
>
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96802
Uroš Bizjak changed:
What|Removed |Added
CC||hjl.tools at gmail dot com
Target
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96793
--- Comment #6 from Uroš Bizjak ---
Ehm...
2006-10-29 Richard Guenther
* config/i386/i386-protos.h (ix86_expand_floorceil): Declare.
(ix86_expand_floorceildf_32): Likewise.
* config/i386/i386.c
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96744
--- Comment #6 from Uroš Bizjak ---
(In reply to Uroš Bizjak from comment #5)
> However, the patch assumes that avx512vp2intersect implies mavx512dq,
> otherwise there is no direct QImode move from mask register to memory
> available.
This is
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96744
--- Comment #5 from Uroš Bizjak ---
(In reply to Hongtao.liu from comment #2)
> Need to add define_insn for movp2qi/movp2hi?
Yes, this is needed to cover some corner cases. Please see attachment 49114.
However, the patch assumes that
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96744
--- Comment #4 from Uroš Bizjak ---
Created attachment 49114
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=49114=edit
Double-reg mask moves
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96744
--- Comment #3 from Uroš Bizjak ---
Created attachment 49112
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=49112=edit
Retune mask <-> general moves cost
It looks to me that mask <-> general cost is too low, so the compiler now
prefers
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96755
Uroš Bizjak changed:
What|Removed |Added
Last reconfirmed||2020-08-24
Ever confirmed|0
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96536
Uroš Bizjak changed:
What|Removed |Added
Target Milestone|--- |8.5
Status|ASSIGNED
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96654
Uroš Bizjak changed:
What|Removed |Added
Ever confirmed|0 |1
Component|target
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96536
--- Comment #6 from Uroš Bizjak ---
(In reply to Hongtao.liu from comment #1)
> I'm testing patch like
>
>emit_insn ((word_mode == SImode)
> ? gen_incsspsi (reg_255)
> : gen_incsspdi (reg_255));
> -
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96536
Uroš Bizjak changed:
What|Removed |Added
Assignee|unassigned at gcc dot gnu.org |ubizjak at gmail dot com
Last
801 - 900 of 6636 matches
Mail list logo