--- Additional Comments From steven at gcc dot gnu dot org 2005-06-06
07:16 ---
Could L1 icache blow-out be the reason?
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19923
--- Additional Comments From rakdver at atrey dot karlin dot mff dot cuni
dot cz 2005-06-06 07:30 ---
Subject: Re: openssl is slower when compiled with gcc 4.0 than 3.3
Could L1 icache blow-out be the reason?
This is not likely with the minimized example.
--
--- Additional Comments From giovannibajo at libero dot it 2005-06-06
13:33 ---
Uhm, at this point, I don't believe anymore that the loop I posted is the cause
of the regression. Maybe the regression is somewhere else. I'll investigate.
--
--- Additional Comments From giovannibajo at libero dot it 2005-06-06
14:40 ---
Looks like the culrpit is this:
=
static unsigned int S[256];
unsigned
md2_block (unsigned int *sp1, unsigned int *sp2, const unsigned
--- Additional Comments From rakdver at atrey dot karlin dot mff dot cuni
dot cz 2005-06-06 15:00 ---
Subject: Re: openssl is slower when compiled with gcc 4.0 than 3.3
Looks like the culrpit is this:
=
--- Additional Comments From rakdver at atrey dot karlin dot mff dot cuni
dot cz 2005-06-02 08:01 ---
Subject: Re: openssl is slower when compiled with gcc 4.0 than 3.3
The assembler attributed to 4.0 was produced by mainline (or some
patched version of 4.0), wasn't it?
Otherwise I
--- Additional Comments From yx at cs dot ucla dot edu 2005-06-01 20:47
---
When we ran 'openssh speed md2', we did see that gcc-4.0 was slower
than earlier versions, so we created a minimal test case, which
we will attach. Here is how long it took to run a 34 megabyte
file through the
--- Additional Comments From pinskia at gcc dot gnu dot org 2005-06-01
20:55 ---
I would not doubt this is just not using the i386 address mode
--
What|Removed |Added
--- Additional Comments From giovannibajo at libero dot it 2005-06-01
22:55 ---
Confirmed. The regression appears only with -fPIC, and it's pretty evident. The
core is md2_block, the inner loop:
GCC 3.4
=
.L29:
xorl
--- Additional Comments From giovannibajo at libero dot it 2005-06-01
22:59 ---
I wonder if this is fixed by TARGET_MEM_REF.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19923
--- Additional Comments From pinskia at gcc dot gnu dot org 2005-05-14
20:36 ---
No feedback in 3 months.
--
What|Removed |Added
Status|WAITING
--
What|Removed |Added
Component|c |target
Keywords||missed-optimization
--- Additional Comments From pinskia at gcc dot gnu dot org 2005-02-12
22:24 ---
We need a self contained example.
--
What|Removed |Added
CC|
13 matches
Mail list logo
