Hi,
I have spent a lot of time last year measuring php performance on
multi-threaded vs single threaded fastcgi.
I had used iplanet web server (instead of apache) for my measurement. Out of
the box single threaded php apps were very close (better) than multi-threaded
versions but when we started o
Hi,
On non x86_64 platforms current implementation of safe_address uses doubles
for 64 bit php which is very expensive on some architectures e.g sparc.
safe_address function calculates nmemb * size + offset and it want to make
sure that it output doesn't overflow. My colleague Richard Smith su
On Thu, Nov 12, 2009 at 11:54:55AM -0800, Rasmus Lerdorf wrote:
> A couple of notes.
>
> You make it sound like this happens on all includes. It is only
> include_once/require_once that have this problem. Regular
> include/require do not.
Sorry for making it confusing. I meant only for include_o
Hi,
When php script includes a script, php engine compiles the included file.
When opcode caching extensions e.g apc are used, re-compilation is avoided by
apc. But there is a performance problem here. Since php code uses
zend_stream_open to open the included file, open/close system call is s
+1
Regards,
Basant.
On Thu, Nov 05, 2009 at 05:22:34PM +0530, Arvind Srinivasan wrote:
> When compiled for multi-threaded (#ifdef ZTS ) operation, the various
> subsystems in PHP use dynamically allocated (ts_allocate_id)
> identifiers to index into the thread-local storage for each subsystem.
>
I have revised the patch for the race condition. New patch is much
less invasive and
scope of the change is restricted to pdo only.
Details are there in the bug.
Patch is attached.
Regards,
Basant.
On Tue, Oct 20, 2009 at 4:03 PM, Basant Kukreja
wrote:
> Hi,
> There is a race condit
Hi,
There is a race condition in pdo's
PDOStatement->ce.default_properties.ref_count. The integer
is incremented without any lock around it (or using any other atomic APIs).
This causes PDO to crash under stress. Details are given in bug report
http://bugs.php.net/bug.php?id=49937&thanks=1
I h
On Aug 16, 2009, at 2:25 PM, Paul Biggar wrote:
I've seen that you talked about branch prediction misses in your
paper. Did you ever tried to compile the original PHP with different
compilers, for example from intel or sun studio? Did you test or have
you ever heard of someone who tested the
I have measured Quercus php performance compared to standard php with
APC on a ecommerce workload and found that Quercus php performed
slower than standard php + APC.
Regards,
Basant.
On Aug 16, 2009, at 11:16 AM, Tom Boutell wrote:
Re: the performance of PHP, if it's badly implemented, should
