Where would you want to start UPGRADING from?
The main problem I faced with the 5.1.0 version was not knowing how far back
to go... or what to count 'in'. E.g. something that was only in CVS for a
few weeks and no releases - should that be included?
Needs some thought for 6.0
- Orig
If this is possible, like Dmitry said, with a macro, that would be
interesting. I'm curious what this macro would look like. Personally, I tend
to go for the less readable solution if the performance advantage is
definately there. A little comment to the code should always be able to
clearify i
Pierre and all,
A while ago I suggested to add an UPGRADING file for the PHP 6
upgrade, where we'd document every time we break BC and how to solve
that. For example, with nuking register_globals we could write what
we broke, and if really needed, what 1-2 liners they can write to get
the sam
I don't think it gets called enough to be substantial and worth
making this harder to maintain (unless I'm proven wrong :)
Andi
At 10:51 PM 3/13/2006, Dmitry Stogov wrote:
In case if this routine is really called to frequency, we can use readable
macros for this optimization.
Thanks. Dmitry.
In case if this routine is really called to frequency, we can use readable
macros for this optimization.
Thanks. Dmitry.
> -Original Message-
> From: Andi Gutmans [mailto:[EMAIL PROTECTED]
> Sent: Tuesday, March 14, 2006 8:57 AM
> To: Marcus Boerger; Ron Korving
> Cc: internals@lists.ph
This one isn't a good idea. I bet it won't affect overal performance
but makes the code much less maintainable.
The others look OK (just took a quick glance)
At 01:20 PM 3/13/2006, Marcus Boerger wrote:
> -
> In sapi/apache/mod_php5.c in the function php_apache_flag_handler_ex(),
> the or
Hello Ron,
i was under the impression you were already contributing to one of the
extensions. And well 5.1.2 obviously differs pretty much from 5.1.3 right
now especially in that file. To apply for a cvs account you'd first send
in some patches and mostly you would do that for a certain extensio
In zend.h, ZSTR is defined like this:
typedef union _zstr {
char *s;
UChar *u;
void *v;
} zstr;
#ifdef __GNUC__
# define ZSTR(x)((zstr)(x))
# define NULL_ZSTR ZSTR((void*)NULL)
# define EMPTY_ZSTR ZSTR("\0\0")
#else
extern ZEND_API zstr null_zstr;
Hi Marcus,
Actually, I was looking at the 5.1.2, not 4.4. And 5.1.2 can't differ that
much from 5.1/HEAD, right? I'm not sure if I'm ready for CVS access, since I
don't know enough of the architecture of the system as a whole. I wouldn't
wanna break anything while trying to make things better. If
Hey,
Thanks for posting this info. It definitely sounds like we should
concentrate on the 0 length script at this point. I saw Dmitry
already made some good improvements.
It'd be helpful if others also run such an empty benchmark because it
seems like the two trees are on par now and that it
Hello Ron,
just as a clarification, you looked at unchanged 4.4 code that is fixed
since long in 5.1/HEAD. Please always first look into 5.1/HEAD since 4.4
will only get real fixes but no code beautifying. Also we always start to
modify HEAD first and MFH from there. Doing it the otherway round
Hi,
I don't have access and I don't need to do the changes myself, but if you
prefer it, I will (provided I get access of course).
Ron
"Marcus Boerger" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]
> Hello Ron,
>
> that stuff is only used in edgcases however it is more of a fix
FWIW, I support this motion.
I am never a fan of the software lifecycle that looks like
"here's a useful technology, now we just need this,
and this...etc. ad nausea". Why?
(1) Adds complexity
(2) You often get pulled out of your original design philosophy
which puts the code's architectu
Hello Ron,
that stuff is only used in edgcases however it is more of a fix than an
optimization. Do you have access and want to do the changes yourself?
regards
marcus
Monday, March 13, 2006, 10:08:30 PM, you wrote:
> Hi,
> If you're even interested in the tinyest of optimizations, you may w
Hi,
If you're even interested in the tinyest of optimizations, you may wanna
read this. I was just going through the php code. I'm not familiar with it
at all, so I don't know which functions are the bottlenecks, so I can't help
in optimizing the big picture. But I had little else to do right now
Andi Gutmans wrote:
Thanks for posting this info. It definitely sounds like we should
concentrate on the 0 length script at this point. I saw Dmitry already
made some good improvements.
Yup, that patch helped. And I guess on some architectures 5.1 is faster
now, but there is still a bit of a
Hello Dmitry,
if you mean the hash stuff you changed then you did quite some mistakes.
Because the normal apply functions don't respect the ZEND_HASH_* consts as
i mailed last week.
marcus
Monday, March 13, 2006, 1:12:01 PM, you wrote:
> Hi Rasmus,
> I made two improvements in 5.1 and run th
The difference is probably due to the big cache that Centrino has. So the
problem with AMD can be a high number of cache-misses. Try to compile with a
recent compiler (aka gcc 4) with -Os, to see if it improves.
I do also think that the Zend Engine may be inlining too many functions. The
kernel
Like case-sensitivity.
-Andrei
P.S. Joking, I am _just_ JOKING! Have a sense of humor, people.
On Mar 12, 2006, at 8:50 AM, Pierre wrote:
On 3/12/06, Steph Fox <[EMAIL PROTECTED]> wrote:
Lukas, that's extremely odd. Check history! (when most of the
core team
agreed on jump originally, fo
C++ Templates: The complete guide (David Vandevoorde)
addresses the issue.
First workaround is simply to include what would normally be
the .cpp code in the .h header. Not sure that's ideal, but it works,
so that's the issue resolved, (or sidelined for now).
Apparently there's also export, but fi
The saga continues, with progress towards a (hoped for) marriage
between Php and C++ that 'looks like' C++ to the extension developer.
I'm exploring a templated version of PhpClassFactory (vs possible root
PhpClassFactoryObject) which is making progress, but a linking error
not dissimilar to sepa
On 13.03.2006 18:05, Nick Mitin wrote:
I was trying to create 2 mysql connections.
$l1 = mysql_pconnect(“localhost”, “user”, “pass”);
mysql_select_db(“db1”, $l1);
$l2 = mysql_pconnect(“localhost”, “user”, “pass”);
mysql_select_db(“db1”, $l2);
The $l1 and $l2 now contains different resource ids
Nick Mitin wrote:
I was trying to create 2 mysql connections.
this is a user question and should therefore not be directed at the
internals list..
1) persistent connections have issues:
http://de2.php.net/manual/en/features.persistent-connections.php
2) use the new link parameter to get a n
I was trying to create 2 mysql connections.
$l1 = mysql_pconnect(“localhost”, “user”, “pass”);
mysql_select_db(“db1”, $l1);
$l2 = mysql_pconnect(“localhost”, “user”, “pass”);
mysql_select_db(“db1”, $l2);
The $l1 and $l2 now contains different resource ids (#1, #2), but
mysql_query(“select * fro
Yes, fresh CVS update of 4.4, 5.1 and apc.
Edin
Dmitry Stogov wrote:
> Have you rebuilt 5.1 HEAD today?
>
> Dmitry.
>
>
>>-Original Message-
>>From: Edin Kadribasic [mailto:[EMAIL PROTECTED]
>>Sent: Monday, March 13, 2006 5:32 PM
>>To: Dmitry Stogov
>>Cc: 'Rasmus Lerdorf'; 'internals
Have you rebuilt 5.1 HEAD today?
Dmitry.
> -Original Message-
> From: Edin Kadribasic [mailto:[EMAIL PROTECTED]
> Sent: Monday, March 13, 2006 5:32 PM
> To: Dmitry Stogov
> Cc: 'Rasmus Lerdorf'; 'internals'
> Subject: Re: [PHP-DEV] Calling performance geeks
>
>
> I also see very little
I also see very little difference (in favor of php4) on my test box
(Dual Xeon 3.2GHz, running Linux 2.6.12 Fedora Core 3):
php-5.1 (top.php)
plain 3723 req/sec
apc stat=1 6220 req/sec
apc stat=0 6278 req/sec
php-4.4 (4top.php)
plain 3978 req/sec
apc stat=1 6421 req/sec
apc stat=0 6650 req/sec
p
Hi Rasmus,
I made two improvements in 5.1 and run the same bechmarks on Intel Pentium M
1.5GHz 2M cache.
top/top5/top10
php-5.1 740 550 430 req/sec
php-4.4 680 440 290 req/sec
May be the problem is AMD chip? :)
Thanks. Dmitry.
> -Original Me
Marcus Boerger wrote:
Hello Rasmus,
not a thing for 5.1 or 4.4 but in 5.2 we could change to a case
insensitive comparison function. That would allow us to change nearly
all of strcasecmp to memcmp. And in may cases it means one less
allocation. And it also means a lot of less code. The casins
Hello Rasmus,
not a thing for 5.1 or 4.4 but in 5.2 we could change to a case
insensitive comparison function. That would allow us to change nearly
all of strcasecmp to memcmp. And in may cases it means one less
allocation. And it also means a lot of less code. The casinsensitive
comparision is
Hello,
that's by design, we don't want to use strlen on ever execution. Instead the
macro use sizeof() which is done at compile time. The workaround would be to
set the string lentgh after using the macro which works pretty well.
Sidenote, you should be using 'const char* var = "string";' or even
PHP 5 Bug Database summary - http://bugs.php.net
Num Status Summary (458 total including feature requests)
===[*Compile Issues]==
36404 Open configure script cannot complete libxml build
===
On Sat, 11 Mar 2006, Janwillem Borleffs wrote:
> Translating the PHP documentation to Dutch.
Can you approve this one please?
regards,
Derick
--
PHP Internals - PHP Runtime Development Mailing List
To unsubscribe, visit: http://www.php.net/unsub.php
PHP 4 Bug Database summary - http://bugs.php.net
Num Status Summary (632 total including feature requests)
===[*Configuration Issues]
36257 Open php ini master values are reset between vhosts
==
34 matches
Mail list logo
