Re: perfomance and regular expressions
Gary W. Swearingen wrote: Terry Lambert [EMAIL PROTECTED] writes: Gary W. Swearingen wrote: Anybody know any languages that allow compile-time (and/or link-time) computations using (most of?) the same language? I've often desired the feature. (I suppose some preprocessor like m4 could handle some of it.) LISP. Forth. Smalltalk. Prolog. BASIC. Oh yeah... Java. Can any of those execute or interpret some code and then compile the results of that into a non-interpreted executable? Yes. All of them. Maybe I HAVE heard of some interpreters (Python?) saving an excecutable image of themselves after running a while. Is that what you're thinking of? Will all of those language systems above do that? It's not quite what I was thinking of, but I guess it meets my (poor) specifications. Franz LISP used to do this; EMACS did, as well. So did VAX LISP. The command was undump for EMACS. It's still there. The point was to send a SIGABRT, get a coredump, and then turn it into an executable that could be reloaded with the precompiled context intact. One of the earliest relaxations that was a tradeoff between security and usability was not enforcing the file size match the a.out headers (e.g. the file could be larger than the a,.out headers indicated) to permit things like pre-compiled code to be appended to the end of an image file, so that it wouldn't have to be recompiled the next time the image ran. There's a long and glorious history of generating code incrementally in an otherwise interpreted language. Java didn't invengt anything new when it invented bytecode and JIT... mostly, it just reinvented the UCSD P-code system from the mid/early 1970's. I can probably list languages you've never heard of which have the same attribute, and that some people on this list have actually coded in ;^). Anything that supports a JIT, by definition, is capable of doing what you suggest. Don't they teach History of Computing to people any more?!? -- Terry To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: perfomance and regular expressions
On Sat, Mar 30, 2002 at 01:39:14AM -0800, Terry Lambert wrote: Gary W. Swearingen wrote: Terry Lambert [EMAIL PROTECTED] writes: Gary W. Swearingen wrote: Anybody know any languages that allow compile-time (and/or link-time) computations using (most of?) the same language? I've often desired the feature. (I suppose some preprocessor like m4 could handle some of it.) LISP. Forth. Smalltalk. Prolog. BASIC. Oh yeah... Java. ... There's a long and glorious history of generating code incrementally in an otherwise interpreted language. Java didn't invengt anything new when it invented bytecode and JIT... mostly, it just reinvented the UCSD P-code system from the mid/early 1970's. There was even hardware build that could directly execute p-code. I think the same chips were used (Texas Instruments??) that DEC used to build the LSI11/2 CPUs -- | / o / /_ _ [EMAIL PROTECTED] |/|/ / / /( (_) Bulte Arnhem, the Netherlands We are FreeBSD. Resistance is futile. Prepare to be committed. To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
lang/icc doesn't compile c++ sources
hi hackers, alexander! i've installed the icc port and trying to compile... first of all, C source files compile (and then link with gcc). no problems with that. but i can't get icc to compile a single c++ source: it always fails on the headers. just a simple echo '#includeiostream' comp.cc won't compile. i've tested the include files of o icc o gcc 2.95.3 (the one from my base system, 4.5-S) o gcc 3.0.4 (from the ports) each one gives me a different error. did anybody already solve this problem (by creating new headers?) if so, please tell me! if not, i'd like to work on that, but i don't know where to start. cheerz simon -- /\ http://corecode.ath.cx/ \ / \ ASCII Ribbon Campaign / \ Against HTML Mail and News msg33184/pgp0.pgp Description: PGP signature
Re: lang/icc doesn't compile c++ sources
Simon 'corecode' Schubert wrote:
just a simple echo '#includeiostream' comp.cc
won't compile. i've tested the include files of
o icc
o gcc 2.95.3 (the one from my base system, 4.5-S)
o gcc 3.0.4 (from the ports)
each one gives me a different error.
did anybody already solve this problem (by creating new headers?)
if so, please tell me!
if not, i'd like to work on that, but i don't know where to start.
Try this patch:
--- bad Sat Mar 30 05:18:02 2002
+++ goodSat Mar 30 05:18:23 2002
@@ -1 +1,2 @@
-echo '#includeiostream' comp.cc
+echo '#include iostream' comp.cc
+echo 'main() {}' comp.cc
8-) 8-) 8-)
-- Terry
To Unsubscribe: send mail to [EMAIL PROTECTED]
with unsubscribe freebsd-hackers in the body of the message
Re: Writing a file system? Docs? Info? Article?
On Thu, 28 Mar 2002, Paolo Pisati wrote: Sometimes ago, I heard someone wanted to write a fs example (article? howto?) just to teach how to write a real fs under FreeBSD. I'm looking for this kind of info, any good pointer is welcome... Unfortunately, as others have pointed out, there aren't really such documents, and the best reference is the source code. That said, some file system source makes for a better read than other source :-). I actually found cd9660 to be a relatively readable starting point for a read-only filesystem. UFS is a useful reference in the sense that if performance is a goal, UFS knows about the intensely evil relationship a filesystem can have with the VM system, but otherwise the complexity of UFS can be a bit off-putting. The UDF patches floating around also make for a very straight-forward read-only fs implementation--the source code is small, simple, and concise. When it comes to writable filesystems, the pain threshold increases quite a bit, and you might want to look at one of the other local filesystems such as HPFS. If you're writing a distributed filesystem, there are lots of decent examples of that, especially if you use local caching objects to hold the data so that you don't have to much with the buffer cache (for example, Arla and Coda do this). In terms of general reference, the first challenge is getting a grasp of VFS. For this, the The Design and Implementation of 4.4 BSD by McKusick, Et Al is always a good start. I found that it was also useful to read up on the VFS stuff in Solaris Internals, and Unix Internals. If you know what's good for you, you'll avoid stacked file systems at any cost. :-) One of the most useful sources of vop information is actually src/sys/kern/vnode_if.src, which contains the argument lists and locking protocol. The locking protocol isn't strictly held to in a number of cases, which can cause confusion, but it's a good reference to have in hand. Good luck. When you discover bugs in our VFS implementation, please make sure to file bug reports :-). Robert N M Watson FreeBSD Core Team, TrustedBSD Project [EMAIL PROTECTED] NAI Labs, Safeport Network Services To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: GPS time.
In a message written on Fri, Mar 29, 2002 at 06:04:11PM -0600, Paul Halliday wrote: I just connected my gps (garmin gps III plus) to my serial port and realized that simply cat'ing cua0 displays date/time/position of the unit. (neato). Anyway, how accurate would it be to use the time from this output for ntp as opposed to my current setup using ntp servers. Your NTP servers are better. I tested a III Plus, and without a 1 PPS source (which that model doesn't provide) it's accurate to about 100ms, give or take. Since real NTP servers are 1ms, they really aren't that good. It's not that the time isn't accurate, it's that they were not designed to communicate with that accuracy to an external device. If you NTP off the Internet, and want a local backup clock it might be an acceptable solution. However clocks that can achieve 1ms accuracy can be had for $1000, so if you really care you should get one of those. You might want to do some searches for NTP in google. -- Leo Bicknell - [EMAIL PROTECTED] - CCIE 3440 PGP keys at http://www.ufp.org/~bicknell/ Read TMBG List - [EMAIL PROTECTED], www.tmbg.org To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: lang/icc doesn't compile c++ sources
On Sat, 30 Mar 2002 05:16:38 -0800 Terry Lambert [EMAIL PROTECTED] wrote:
Simon 'corecode' Schubert wrote:
just a simple echo '#includeiostream' comp.cc
won't compile. i've tested the include files of
o icc
o gcc 2.95.3 (the one from my base system, 4.5-S)
o gcc 3.0.4 (from the ports)
each one gives me a different error.
did anybody already solve this problem (by creating new headers?)
if so, please tell me!
if not, i'd like to work on that, but i don't know where to start.
Try this patch:
--- bad Sat Mar 30 05:18:02 2002
+++ goodSat Mar 30 05:18:23 2002
@@ -1 +1,2 @@
-echo '#includeiostream' comp.cc
+echo '#include iostream' comp.cc
+echo 'main() {}' comp.cc
8-) 8-) 8-)
;] no, that's definively not the problem.
as alexander wrote, there are several errors in cwchar etc...
go ahead and try yourself if you don't believe me...
cheerz
simon
--
/\ http://corecode.ath.cx/
\ /
\ ASCII Ribbon Campaign
/ \ Against HTML Mail and News
msg33188/pgp0.pgp
Description: PGP signature
Re: lang/icc doesn't compile c++ sources
On 30 Mär, Simon 'corecode' Schubert wrote:
but i can't get icc to compile a single c++ source:
it always fails on the headers.
just a simple echo '#includeiostream' comp.cc
won't compile. i've tested the include files of
o icc
o gcc 2.95.3 (the one from my base system, 4.5-S)
o gcc 3.0.4 (from the ports)
each one gives me a different error.
I tried this:
--snip---
#includeiostream
int main(void)
{
cout Hello World;
exit(0);
}
---snip---
- gcc from -current didn't fails (and no errors).
- icc complains about:
---snip---
test.cpp:
/usr/include/stdlib.h, line 57: error: invalid combination of type specifiers
typedef _BSD_WCHAR_T_ wchar_t;
^
/usr/local/intel/compiler50/ia32/include/cwchar, line 16: error: the global scope
has no tm
using ::mbstate_t; using ::size_t; using ::tm; using ::wint_t;
^
/usr/local/intel/compiler50/ia32/include/cwchar, line 19: error: the global scope
has no btowc
using ::btowc;
^
/usr/local/intel/compiler50/ia32/include/cwchar, line 22: error: the global scope
has no mbrlen
using ::mbrlen; using ::mbrtowc; using ::mbsrtowcs;
^
[and so on]
---snip---
I think this is because of our wchar implementation (mostly a stub if I
remember correctly), but I could be wrong on this.
if not, i'd like to work on that, but i don't know where to start.
If you have to patch some Intel headers, then you have to work on a
make pre-patch version of icc. Don't forget to apply patch-include by
hand before you make diffs.
Bye,
Alexander.
--
The best things in life are free, but the
expensive ones are still worth a look.
http://www.Leidinger.net Alexander @ Leidinger.net
GPG fingerprint = C518 BC70 E67F 143F BE91 3365 79E2 9C60 B006 3FE7
To Unsubscribe: send mail to [EMAIL PROTECTED]
with unsubscribe freebsd-hackers in the body of the message
Re: GPS time.
In a message written on Fri, Mar 29, 2002 at 06:04:11PM -0600, Paul Halliday wrote: I just connected my gps (garmin gps III plus) to my serial port and realized that simply cat'ing cua0 displays date/time/position of the unit. (neato). Anyway, how accurate would it be to use the time from this output for ntp as opposed to my current setup using ntp servers. Your NTP servers are better. If you want to buy an appropriate GPS for this, you may wish to check the Tucson Amateur Packet Radio site (www.tapr.org) and look at the Motorola Oncore UT+ board. For around $200 you can have a stratum 1 quality time reference. Regards, -Les -- Les Biffle (480) 585-4099[EMAIL PROTECTED] http://www.les.safety.net/ Network Safety Corp., 5831 E. Dynamite Blvd., Cave Creek, AZ 85331 To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: perfomance and regular expressions
Terry Lambert [EMAIL PROTECTED] writes: Don't they teach History of Computing to people any more?!? They didn't offer it to me when I was studing for my BSEE in the 70s. (And I haven't been paying nearly as much attention as you have.) I hate to think what my alma mata is teaching for HoC after getting several multi-million dollar gifts from Bill Gates and Company. To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: perfomance and regular expressions
On Sat, Mar 30, 2002 at 10:22:37AM -0800, Gary W. Swearingen wrote: Terry Lambert [EMAIL PROTECTED] writes: Don't they teach History of Computing to people any more?!? They didn't offer it to me when I was studing for my BSEE in the 70s. (And I haven't been paying nearly as much attention as you have.) I hate to think what my alma mata is teaching for HoC after getting several multi-million dollar gifts from Bill Gates and Company. Simple: The Internet, anothher major M$ Invention. Bah! -- | / o / /_ _ [EMAIL PROTECTED] |/|/ / / /( (_) Bulte Arnhem, the Netherlands We are FreeBSD. Resistance is futile. Prepare to be committed. To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: GPS time.
http://www.gpsclock.com/ is $380US and does PPS pulses accurate to plus or minus 1 microsecond of UTC. On Sat, Mar 30, 2002 at 09:28:59AM -0500, Leo Bicknell wrote: In a message written on Fri, Mar 29, 2002 at 06:04:11PM -0600, Paul Halliday wrote: I just connected my gps (garmin gps III plus) to my serial port and realized that simply cat'ing cua0 displays date/time/position of the unit. (neato). Anyway, how accurate would it be to use the time from this output for ntp as opposed to my current setup using ntp servers. Your NTP servers are better. I tested a III Plus, and without a 1 PPS source (which that model doesn't provide) it's accurate to about 100ms, give or take. Since real NTP servers are 1ms, they really aren't that good. It's not that the time isn't accurate, it's that they were not designed to communicate with that accuracy to an external device. If you NTP off the Internet, and want a local backup clock it might be an acceptable solution. However clocks that can achieve 1ms accuracy can be had for $1000, so if you really care you should get one of those. You might want to do some searches for NTP in google. -- Leo Bicknell - [EMAIL PROTECTED] - CCIE 3440 PGP keys at http://www.ufp.org/~bicknell/ Read TMBG List - [EMAIL PROTECTED], www.tmbg.org To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: lang/icc doesn't compile c++ sources
Simon 'corecode' Schubert wrote:
Try this patch:
--- bad Sat Mar 30 05:18:02 2002
+++ goodSat Mar 30 05:18:23 2002
@@ -1 +1,2 @@
-echo '#includeiostream' comp.cc
+echo '#include iostream' comp.cc
+echo 'main() {}' comp.cc
8-) 8-) 8-)
;] no, that's definively not the problem.
as alexander wrote, there are several errors in cwchar etc...
go ahead and try yourself if you don't believe me...
I did: g++ complained about the lack of the space, and then
complained about the lack of a main(). The patch fixed all
the g++ complaints.
-- TErry
To Unsubscribe: send mail to [EMAIL PROTECTED]
with unsubscribe freebsd-hackers in the body of the message
Re: perfomance and regular expressions
Gary W. Swearingen wrote: Terry Lambert [EMAIL PROTECTED] writes: Don't they teach History of Computing to people any more?!? They didn't offer it to me when I was studing for my BSEE in the 70s. (And I haven't been paying nearly as much attention as you have.) I hate to think what my alma mata is teaching for HoC after getting several multi-million dollar gifts from Bill Gates and Company. Americans invented computers C# is the first language to support compiled code with a runtime Kerberos is *supposed* to need an Active Directory server to work ... -- Terry To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Do you sell cabinets?
If you sell, install or make cabinets, take a moment. This product is well worth looking into. 1. Reduce damage to cabinets tremendously; 2. The faces of the cabinets will be flush, the tops and bottoms even; 3. Cuts installation time by 1/2 or more with less fatigue; 4. One man cabinet installation with Micro Adjusting accuracy. These are only a few benefits. Quality workmanship reflects: Your name Your image Your product Remember, the women always look at the kitchen. For the discriminate cabinet maker and installer. http://www.quicksitemaker.com/members/bbiles If you are a cabinet manufacturer your distributers will benefit. Cabinet Eaz makes the difference. To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: lang/icc doesn't compile c++ sources
--- bad Sat Mar 30 05:18:02 2002
+++ goodSat Mar 30 05:18:23 2002
@@ -1 +1,2 @@
-echo '#includeiostream' comp.cc
+echo '#include iostream' comp.cc
+echo 'main() {}' comp.cc
glibness aside, this doesn't work either:
flapjack:/home/nick echo '#include iostream' comp.cc
flapjack:/home/nick echo 'main() {}' comp.cc
flapjack:/home/nick icc comp.cc
comp.cc:
/usr/include/stdlib.h, line 57: error: invalid combination of type specifiers
typedef _BSD_WCHAR_T_ wchar_t;
[piles more warnings chopped
This, of course, indicates nothing more than that the linux_devtools
package hasn't been installed: a simple pkg_add -r linux_devtools
appears to fix the problem. The devtools package should be included on
the port dependency list to fix it properly.
Also, the intel folks don't seem to understand csh syntax very well,
according to the last line of compiler50/ia32/bin/iccvars.csh, which
states:
setenv INTEL_FLEXLM_LICENSE=INSTALLDIR/licenses
Oops - that should have been caught by QA. Unfortunately, this file is
modified by pre-patch:, which means that a simple patch won't do the
trick if people decide to install the port in a different location.
Hmmm.
I've mailed gnats about both of these problems, and have included
suggested fixes for each.
Nick
To Unsubscribe: send mail to [EMAIL PROTECTED]
with unsubscribe freebsd-hackers in the body of the message
Re: lang/icc doesn't compile c++ sources
Nick Hilliard wrote:
--- bad Sat Mar 30 05:18:02 2002
+++ goodSat Mar 30 05:18:23 2002
@@ -1 +1,2 @@
-echo '#includeiostream' comp.cc
+echo '#include iostream' comp.cc
+echo 'main() {}' comp.cc
glibness aside, this doesn't work either:
Works with g++... though I expect glib-ness... ;^)...
/usr/include/stdlib.h, line 57: error: invalid combination of type specifiers
typedef _BSD_WCHAR_T_ wchar_t;
This seems to be a problem with the wchar_t being intrinsic,
and defined anyway for the compiler.
[piles more warnings chopped
This, of course, indicates nothing more than that the linux_devtools
package hasn't been installed: a simple pkg_add -r linux_devtools
appears to fix the problem. The devtools package should be included on
the port dependency list to fix it properly.
Actually, when you install the Linux devtools, then what you
get is a Linux binary search path inserted first. This is
actually the *wrong* thing, if what you want is to use the
compiler to generate native FreeBSD binaries.
Probably, it would be best to handle the warnings, each on a
case-by-case basis.
Unless what you really want is to generate Linux binaries on
FreeBSD?
Also, the intel folks don't seem to understand csh syntax very well,
according to the last line of compiler50/ia32/bin/iccvars.csh, which
states:
setenv INTEL_FLEXLM_LICENSE=INSTALLDIR/licenses
Oops - that should have been caught by QA. Unfortunately, this file is
modified by pre-patch:, which means that a simple patch won't do the
trick if people decide to install the port in a different location.
Hmmm.
I've mailed gnats about both of these problems, and have included
suggested fixes for each.
I think this is just a sed path that is supposed to be replaced
during the installation. This is a probably bug in the FreeBSD
install replacement, rather than an Intel bug.
-- Terry
To Unsubscribe: send mail to [EMAIL PROTECTED]
with unsubscribe freebsd-hackers in the body of the message
icc status on FreeBSD
I was just looking into the icc port on FreeBSD. It tells on freshports site that it does not produce native executables but objects and linux compatibility base is necessary. Is it possible (with cooperation with intel) to completely port icc to FreeBSD (and make it compatible with libc so that it will be possible to compile FreeBSD kernel sometime. I was asking the same question to intel guys (through premier.intel.com which you can register for free) and they haven't heard about FreeBSD port. I don't personally care how open source anything is, if it really produces extraordinary machine code in the end. With version 6.0 of icc one would be able to compile linux kernel. I hope with some cooperation one could do the same with FreeBSD. I respect the gcc effort, but if icc is as good as it is told to be then it could supersede (at least in intel community) gcc. What is the general sentiment and the will for cooperation about porting icc to FreeBSD? Alp To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: icc status on FreeBSD
I respect the gcc effort, but if icc is as good as it is told to be then it could supersede (at least in intel community) gcc. What is the general sentiment and the will for cooperation about porting icc to FreeBSD? I doubt that porting will be necessary; I've watched another developer (who will probably come forward when he's ready) compiling large portions of the FreeBSD codebase with the Linux icc binary. Initial performance results were solid but not astounding; we'll see when it's ready, eh? = Mike -- To announce that there must be no criticism of the president, or that we are to stand by the president, right or wrong, is not only unpatriotic and servile, but is morally treasonable to the American public. - Theodore Roosevelt To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: icc status on FreeBSD
I wouldn't necessarily get too excited about using icc for compiling operating systems. The company I work for uses gcc for compiling its file server OS. People their have, however, experimented with using icc for builds. When they used feedback profiling from SFS i.e. to determine which branches are likely taken, (all standard benchmark disclaimers apply) they saw a 17% performance increase. As of 3.0.3 gcc also supports feedback profiling, and using it they also saw a 17% performance increase over a stock gcc compile. Without optimization the stock icc build performs worse than the gcc build. Your mileage may of course vary. My point is just that icc does not appear to perform any wonders for systems code. -Kip --- Alp ATICI [EMAIL PROTECTED] wrote: I was just looking into the icc port on FreeBSD. It tells on freshports site that it does not produce native executables but objects and linux compatibility base is necessary. Is it possible (with cooperation with intel) to completely port icc to FreeBSD (and make it compatible with libc so that it will be possible to compile FreeBSD kernel sometime. I was asking the same question to intel guys (through premier.intel.com which you can register for free) and they haven't heard about FreeBSD port. I don't personally care how open source anything is, if it really produces extraordinary machine code in the end. With version 6.0 of icc one would be able to compile linux kernel. I hope with some cooperation one could do the same with FreeBSD. I respect the gcc effort, but if icc is as good as it is told to be then it could supersede (at least in intel community) gcc. What is the general sentiment and the will for cooperation about porting icc to FreeBSD? Alp To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message __ Do You Yahoo!? Yahoo! Greetings - send holiday greetings for Easter, Passover http://greetings.yahoo.com/ To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: Getting more system information
On Fri, 29 Mar 2002, Ron Chen wrote: --- Baldur Gislason [EMAIL PROTECTED] wrote: See top(1), either parse the output from that in batch mode, or use the source luke, the source! Baldur In fact, I looked for the source before I asked. But I need it in tar.gz format, while the source files on ftp.freebsd.org has something different. Yeah, a split tar.gz format. 'cat swhatever.* | tar xzf -' You might find http://cvsweb.freebsd.org/ more useful. Doug White| FreeBSD: The Power to Serve [EMAIL PROTECTED] | www.FreeBSD.org To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
Re: lang/icc doesn't compile c++ sources
On Sat, 30 Mar 2002 12:17:17 -0800 Terry Lambert [EMAIL PROTECTED] wrote:
Simon 'corecode' Schubert wrote:
Try this patch:
--- bad Sat Mar 30 05:18:02 2002
+++ goodSat Mar 30 05:18:23 2002
@@ -1 +1,2 @@
-echo '#includeiostream' comp.cc
+echo '#include iostream' comp.cc
+echo 'main() {}' comp.cc
8-) 8-) 8-)
;] no, that's definively not the problem.
as alexander wrote, there are several errors in cwchar etc...
go ahead and try yourself if you don't believe me...
I did: g++ complained about the lack of the space, and then
complained about the lack of a main(). The patch fixed all
the g++ complaints.
huh. mine does not:
$ echo '#includeiostream'comp.cc
$ gcc -c comp.cc
$
but that's not the point ;] icc will desperately fail on this because
the header files don't fit.
the headers coming with icc fit icc very well (of course) but don't fit
our system headers and the libs (some functions are being used in
templates which are not [yet?] implemented in our libc)
on the other hand, when i try using the g++ headers, icc will fail on
most gcc'isms that it doesn't understand.
i'm lost, really. which headers should i begin with and patch?
cheerz
simon
--
/\ http://corecode.ath.cx/
\ /
\ ASCII Ribbon Campaign
/ \ Against HTML Mail and News
msg33203/pgp0.pgp
Description: PGP signature
Re: lang/icc doesn't compile c++ sources
Simon 'corecode' Schubert wrote: but that's not the point ;] icc will desperately fail on this because the header files don't fit. the headers coming with icc fit icc very well (of course) but don't fit our system headers and the libs (some functions are being used in templates which are not [yet?] implemented in our libc) on the other hand, when i try using the g++ headers, icc will fail on most gcc'isms that it doesn't understand. i'm lost, really. which headers should i begin with and patch? Perhaps you should install *ONLY* the compiler, and *NOT* install header files for an OS other than FreeBSD, expecting them to work? There was a thread on -current, where a home page of a patch set was posted several times over for getting FreeBSD to compile all of user space with ICC. The user space compilation includes the tiny amount of C++ code actually used in FreeBSD proper. -- Terry To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message
