The current smp branch builds and passes make check on the Raspberry Pi 3 with
17 17:36 git clone https://github.com/gambit/gambit.git 18 17:51 cd gambit 19 17:51 git checkout smp 20 17:51 ./configure 23 17:55 make -j1 current-gsc-boot 24 18:29 ./configure --enable-single-host --enable-multiple-threaded-vms 25 18:30 make from-scratch 26 20:57 make check
Afterwards, this worked:
29 21:13 ./configure 'CC=gcc -mcpu=cortex-a53 -mfpu=neon-vfpv4 -mfloat-abi=hard' --enable-single-host --enable-multiple-threaded-vms --enable-shared --enable-multiple-versions 30 21:15 make mostlyclean 31 21:15 make 32 21:34 make check
Brad
Afficher les réponses par date
Very well! Can you also try “make st” (stress tests)? It will compile the usual unit tests and some other long running tests (particularly for the main thread operations).
BTW, what is the result of:
(pretty-print (list processor-count: (##current-vm-processor-count) cpu-count: (##cpu-count)))
Marc
On Mar 19, 2017, at 9:37 PM, Bradley Lucier lucier@math.purdue.edu wrote:
The current smp branch builds and passes make check on the Raspberry Pi 3 with
17 17:36 git clone https://github.com/gambit/gambit.git 18 17:51 cd gambit 19 17:51 git checkout smp 20 17:51 ./configure 23 17:55 make -j1 current-gsc-boot 24 18:29 ./configure --enable-single-host --enable-multiple-threaded-vms 25 18:30 make from-scratch 26 20:57 make check
Afterwards, this worked:
29 21:13 ./configure 'CC=gcc -mcpu=cortex-a53 -mfpu=neon-vfpv4 -mfloat-abi=hard' --enable-single-host --enable-multiple-threaded-vms --enable-shared --enable-multiple-versions 30 21:15 make mostlyclean 31 21:15 make 32 21:34 make check
Brad
I don't know what's happening, but there's no "st" make target (and you see the commands I gave below).
As for the other question:
sweety-pi:~/programs/gambit> gsi Gambit v4.8.8
(pretty-print (list processor-count: (##current-vm-processor-count)
cpu-count: (##cpu-count))) (processor-count: 4 cpu-count: 4)
*** EOF again to exit sweety-pi:~/programs/gambit> gsi -v v4.8.8 20170202123920 armv7l-unknown-linux-gnueabihf "./configure 'CC=gcc -mcpu=cortex-a53 -mfpu=neon-vfpv4 -mfloat-abi=hard' '--enable-single-host' '--enable-multiple-threaded-vms' '--enable-shared' '--enable-multiple-versions'"
Brad
On 03/19/2017 09:43 PM, Marc Feeley wrote:
Very well! Can you also try “make st” (stress tests)? It will compile the usual unit tests and some other long running tests (particularly for the main thread operations).
BTW, what is the result of:
(pretty-print (list processor-count: (##current-vm-processor-count) cpu-count: (##cpu-count)))
Marc
On Mar 19, 2017, at 9:37 PM, Bradley Lucier lucier@math.purdue.edu wrote:
The current smp branch builds and passes make check on the Raspberry Pi 3 with
17 17:36 git clone https://github.com/gambit/gambit.git 18 17:51 cd gambit 19 17:51 git checkout smp 20 17:51 ./configure 23 17:55 make -j1 current-gsc-boot 24 18:29 ./configure --enable-single-host --enable-multiple-threaded-vms 25 18:30 make from-scratch 26 20:57 make check
Afterwards, this worked:
29 21:13 ./configure 'CC=gcc -mcpu=cortex-a53 -mfpu=neon-vfpv4 -mfloat-abi=hard' --enable-single-host --enable-multiple-threaded-vms --enable-shared --enable-multiple-versions 30 21:15 make mostlyclean 31 21:15 make 32 21:34 make check
Brad
I don't know what's happening, but there's no "st" make target (and you see the commands I gave below).
As for the other question:
sweety-pi:~/programs/gambit> gsi Gambit v4.8.8
(pretty-print (list processor-count: (##current-vm-processor-count)
cpu-count: (##cpu-count))) (processor-count: 4 cpu-count: 4)
*** EOF again to exit sweety-pi:~/programs/gambit> gsi -v v4.8.8 20170202123920 armv7l-unknown-linux-gnueabihf "./configure 'CC=gcc -mcpu=cortex-a53 -mfpu=neon-vfpv4 -mfloat-abi=hard' '--enable-single-host' '--enable-multiple-threaded-vms' '--enable-shared' '--enable-multiple-versions'"
Brad
On 03/19/2017 09:43 PM, Marc Feeley wrote:
Very well! Can you also try “make st” (stress tests)? It will compile the usual unit tests and some other long running tests (particularly for the main thread operations).
BTW, what is the result of:
(pretty-print (list processor-count: (##current-vm-processor-count) cpu-count: (##cpu-count)))
Marc
On Mar 19, 2017, at 9:37 PM, Bradley Lucier lucier@math.purdue.edu wrote:
The current smp branch builds and passes make check on the Raspberry Pi 3 with
17 17:36 git clone https://github.com/gambit/gambit.git 18 17:51 cd gambit 19 17:51 git checkout smp 20 17:51 ./configure 23 17:55 make -j1 current-gsc-boot 24 18:29 ./configure --enable-single-host --enable-multiple-threaded-vms 25 18:30 make from-scratch 26 20:57 make check
Afterwards, this worked:
29 21:13 ./configure 'CC=gcc -mcpu=cortex-a53 -mfpu=neon-vfpv4 -mfloat-abi=hard' --enable-single-host --enable-multiple-threaded-vms --enable-shared --enable-multiple-versions 30 21:15 make mostlyclean 31 21:15 make 32 21:34 make check
Brad
Things worked well on my (much faster) Ubuntu x86-64 box, so I must have made a mistake. I'll look into it (but it takes a long time on the RPI3).
Brad
On 03/19/2017 09:50 PM, Bradley Lucier wrote:
I don't know what's happening, but there's no "st" make target (and you see the commands I gave below).
Are you in the “smp” branch?
git checkout smp make bootclean etc
Marc
P.S. cute machine name!
On Mar 19, 2017, at 9:50 PM, Bradley Lucier lucier@math.purdue.edu wrote:
I don't know what's happening, but there's no "st" make target (and you see the commands I gave below).
As for the other question:
sweety-pi:~/programs/gambit> gsi Gambit v4.8.8
(pretty-print (list processor-count: (##current-vm-processor-count) cpu-count: (##cpu-count)))
(processor-count: 4 cpu-count: 4)
*** EOF again to exit sweety-pi:~/programs/gambit> gsi -v v4.8.8 20170202123920 armv7l-unknown-linux-gnueabihf "./configure 'CC=gcc -mcpu=cortex-a53 -mfpu=neon-vfpv4 -mfloat-abi=hard' '--enable-single-host' '--enable-multiple-threaded-vms' '--enable-shared' '--enable-multiple-versions'"
Brad
On 03/19/2017 09:43 PM, Marc Feeley wrote:
Very well! Can you also try “make st” (stress tests)? It will compile the usual unit tests and some other long running tests (particularly for the main thread operations).
BTW, what is the result of:
(pretty-print (list processor-count: (##current-vm-processor-count) cpu-count: (##cpu-count)))
Marc
On Mar 19, 2017, at 9:37 PM, Bradley Lucier lucier@math.purdue.edu wrote:
The current smp branch builds and passes make check on the Raspberry Pi 3 with
17 17:36 git clone https://github.com/gambit/gambit.git 18 17:51 cd gambit 19 17:51 git checkout smp 20 17:51 ./configure 23 17:55 make -j1 current-gsc-boot 24 18:29 ./configure --enable-single-host --enable-multiple-threaded-vms 25 18:30 make from-scratch 26 20:57 make check
Afterwards, this worked:
29 21:13 ./configure 'CC=gcc -mcpu=cortex-a53 -mfpu=neon-vfpv4 -mfloat-abi=hard' --enable-single-host --enable-multiple-threaded-vms --enable-shared --enable-multiple-versions 30 21:15 make mostlyclean 31 21:15 make 32 21:34 make check
Brad
On 03/20/2017 08:05 AM, Marc Feeley wrote:
Are you in the “smp” branch?
git checkout smp make bootclean etc
Here's what "worked":
103 22:42 cd gambit 104 22:42 git checkout smp 105 22:43 ./configure 106 22:44 make -j1 current-gsc-boot 107 23:19 ./configure --enable-single-host --enable-multiple-threaded-vms 108 23:20 make from-scratch 109 0:15 make check 110 9:11 make st
"make st" failed with:
../gsi/gsi -:tl,~~bin=../bin,~~lib=../lib,~~include=../include -e '(pretty-print (list processor-count: (##current-vm-processor-count) cpu-count: (##cpu-count)))' (processor-count: 2 cpu-count: 4) ../gsi/gsi -:tl,~~bin=../bin,~~lib=../lib,~~include=../include -f ./run-unit-tests.scm -stress -gsc
[ 53| 0] 35% #####........... 1175.4s 06-thread/tfib-stress.scm ********************* FAILED unit-tests/06-thread/tfib-stress.scm WITH EXIT CODE 0 [148| 1] 100% ################ 2294.1s FAILED 1 UNIT TESTS OUT OF 149 (.7%) makefile:184: recipe for target 'st' failed make[1]: *** [st] Error 1 make[1]: Leaving directory '/home/pi/programs/gambit/tests' makefile:151: recipe for target 'st' failed make: *** [st] Error 2
Marc
P.S. cute machine name!
Nobody was supposed to see that ... uh ... it's my daughter's machine ... uh ...
Ah, who's kidding whom? My other RPI is named cutie-pi. I'm just an old softie.
Brad
-
Bradley Lucier -
Bradley Lucier -
Marc Feeley