On 3/23/2024 11:30 PM, Bonita Montero wrote:
> Am 23.03.2024 um 22:04 schrieb Chris M. Thomasson:
>> On 3/23/2024 9:54 AM, Bonita Montero wrote:
>>> Am 23.03.2024 um 03:34 schrieb Chris M. Thomasson:
>>>> On 12/26/2023 9:06 PM, Bonita Montero wrote:
>>>>> Am 26.12.2023 um 21:24 schrieb Chris M. Thomasson:
>>>>>
>>>>>> So, are you familiar with Intel's early hyper threading problem?
>>>>>> There was false sharing between the ...
>>>>>
>>>>> False sharing can only happen between different cores.
>>>>
>>>> Sigh.
>>>
>>> Why ? Do you think false sharing can happen between the threads
>>> of a single core ?
>>>
>>
>> Sigh again...
>
> On a single core CPU with two thread's there's never false sharing.
>

Whatever you say man... ;^o

Am 24.03.2024 um 20:52 schrieb Chris M. Thomasson:
> On 3/23/2024 11:30 PM, Bonita Montero wrote:
>> Am 23.03.2024 um 22:04 schrieb Chris M. Thomasson:
>>> On 3/23/2024 9:54 AM, Bonita Montero wrote:
>>>> Am 23.03.2024 um 03:34 schrieb Chris M. Thomasson:
>>>>> On 12/26/2023 9:06 PM, Bonita Montero wrote:
>>>>>> Am 26.12.2023 um 21:24 schrieb Chris M. Thomasson:
>>>>>>
>>>>>>> So, are you familiar with Intel's early hyper threading problem?
>>>>>>> There was false sharing between the ...
>>>>>>
>>>>>> False sharing can only happen between different cores.
>>>>>
>>>>> Sigh.
>>>>
>>>> Why ? Do you think false sharing can happen between the threads
>>>> of a single core ?
>>>>
>>>
>>> Sigh again...
>>
>> On a single core CPU with two thread's there's never false sharing.
>>
>
> Whatever you say man... ;^o

You said:

> So, are you familiar with Intel's early hyper threading problem?
There was false sharing between the hyperhtreads.

False sharing is a kind of cache thrashing, but not every cache
thrashing is false sharing. False sharing can only happen between
different cores and the Pentium 4 HT does only have one core.

On 3/24/2024 1:00 PM, Bonita Montero wrote:
> Am 24.03.2024 um 20:52 schrieb Chris M. Thomasson:
>> On 3/23/2024 11:30 PM, Bonita Montero wrote:
>>> Am 23.03.2024 um 22:04 schrieb Chris M. Thomasson:
>>>> On 3/23/2024 9:54 AM, Bonita Montero wrote:
>>>>> Am 23.03.2024 um 03:34 schrieb Chris M. Thomasson:
>>>>>> On 12/26/2023 9:06 PM, Bonita Montero wrote:
>>>>>>> Am 26.12.2023 um 21:24 schrieb Chris M. Thomasson:
>>>>>>>
>>>>>>>> So, are you familiar with Intel's early hyper threading problem?
>>>>>>>> There was false sharing between the ...
>>>>>>>
>>>>>>> False sharing can only happen between different cores.
>>>>>>
>>>>>> Sigh.
>>>>>
>>>>> Why ? Do you think false sharing can happen between the threads
>>>>> of a single core ?
>>>>>
>>>>
>>>> Sigh again...
>>>
>>> On a single core CPU with two thread's there's never false sharing.
>>>
>>
>> Whatever you say man... ;^o
>
> You said:
>
> > So, are you familiar with Intel's early hyper threading problem?
> There was false sharing between the hyperhtreads.
>
> False sharing is a kind of cache thrashing, but not every cache
> thrashing is false sharing. False sharing can only happen between
> different cores and the Pentium 4 HT does only have one core.
>

Yawn...

Bonita Montero <Bonita.Montero@gmail.com> writes:

> Am 11.03.2024 um 18:10 schrieb Tim Rentsch:
>
>> Sounds like you're using 1 bit per number, most of which are
>> wasted. If you use a different encoding the memory requirements
>> can be much smaller. How much memory do you have on the box?
>> If you have 64G you should be able to determine all primes
>> less than 1.5 trillion, using a simple encoding.
>
> I'm omitting even numbers and I handle the number two as a
> special case; that's the fastest solution.
>
>> I've mentioned this encoding before but let me give it again.
>> If numbers are considered mod 30, there are only 8 residues
>> that are not divisible by 2, 3, or 5. The 8 residues are
>> 1, 7, 11, 13, 17, 19, 23, and 29. So a single byte can
>> hold all the information needed for 30 numbers, which means
>>
>> 1500000000000 / 30 = 50000000000
>>
>> which is to say 50 gigabytes should suffice.
>
> Show me the code.

Apparently you have missed the point.

Am 20.04.2024 um 17:35 schrieb Tim Rentsch:
> Bonita Montero <Bonita.Montero@gmail.com> writes:
>
>> Am 11.03.2024 um 18:10 schrieb Tim Rentsch:
>>
>>> Sounds like you're using 1 bit per number, most of which are
>>> wasted. If you use a different encoding the memory requirements
>>> can be much smaller. How much memory do you have on the box?
>>> If you have 64G you should be able to determine all primes
>>> less than 1.5 trillion, using a simple encoding.
>>
>> I'm omitting even numbers and I handle the number two as a
>> special case; that's the fastest solution.
>>
>>> I've mentioned this encoding before but let me give it again.
>>> If numbers are considered mod 30, there are only 8 residues
>>> that are not divisible by 2, 3, or 5. The 8 residues are
>>> 1, 7, 11, 13, 17, 19, 23, and 29. So a single byte can
>>> hold all the information needed for 30 numbers, which means
>>>
>>> 1500000000000 / 30 = 50000000000
>>>
>>> which is to say 50 gigabytes should suffice.
>>
>> Show me the code.
>
> Apparently you have missed the point.

I want to see the code for your idea.

Am 20.04.2024 um 18:34 schrieb Bonita Montero:
> Am 20.04.2024 um 17:35 schrieb Tim Rentsch:
>> Bonita Montero <Bonita.Montero@gmail.com> writes:
>>
>>> Am 11.03.2024 um 18:10 schrieb Tim Rentsch:
>>>
>>>> Sounds like you're using 1 bit per number, most of which are
>>>> wasted.  If you use a different encoding the memory requirements
>>>> can be much smaller.  How much memory do you have on the box?
>>>> If you have 64G you should be able to determine all primes
>>>> less than 1.5 trillion, using a simple encoding.
>>>
>>> I'm omitting even numbers and I handle the number two as a
>>> special case;  that's the fastest solution.
>>>
>>>> I've mentioned this encoding before but let me give it again.
>>>> If numbers are considered mod 30, there are only 8 residues
>>>> that are not divisible by 2, 3, or 5.  The 8 residues are
>>>> 1, 7, 11, 13, 17, 19, 23, and 29.  So a single byte can
>>>> hold all the information needed for 30 numbers, which means
>>>>
>>>>      1500000000000 / 30 = 50000000000
>>>>
>>>> which is to say 50 gigabytes should suffice.
>>>
>>> Show me the code.
>>
>> Apparently you have missed the point.
>
> I want to see the code for your idea.
>

Eh, wij's idea ...

Bonita Montero <Bonita.Montero@gmail.com> writes:

> Am 20.04.2024 um 17:35 schrieb Tim Rentsch:
>
>> Bonita Montero <Bonita.Montero@gmail.com> writes:
[...]
>>> Show me the code.
>>
>> Apparently you have missed the point.
>
> I want to see the code for your idea.

Yes I already understood what you want. That is what
led me to conclude that you have missed the point.

Am 24.04.2024 um 21:28 schrieb Tim Rentsch:
> Bonita Montero <Bonita.Montero@gmail.com> writes:
>
>> Am 20.04.2024 um 17:35 schrieb Tim Rentsch:
>>
>>> Bonita Montero <Bonita.Montero@gmail.com> writes:
> [...]
>>>> Show me the code.
>>>
>>> Apparently you have missed the point.
>>
>> I want to see the code for your idea.
>
> Yes I already understood what you want. That is what
> led me to conclude that you have missed the point.

I don't have "missed the point"; I just want to see the code
basing on the mentioned idea that is faster than my code.

Bonita Montero <Bonita.Montero@gmail.com> writes:

> Am 24.04.2024 um 21:28 schrieb Tim Rentsch:
>
>> Bonita Montero <Bonita.Montero@gmail.com> writes:
>>
>>> Am 20.04.2024 um 17:35 schrieb Tim Rentsch:
>>>
>>>> Bonita Montero <Bonita.Montero@gmail.com> writes:
>>
>> [...]
>>
>>>>> Show me the code.
>>>>
>>>> Apparently you have missed the point.
>>>
>>> I want to see the code for your idea.
>>
>> Yes I already understood what you want. That is what
>> led me to conclude that you have missed the point.
>
> I don't have "missed the point"; [...]

There is more than one school of thought on that question.

I've been playing with this. Has it really been this long? I ought to
have more time than this...

I put together some code, and have been playing with it far too much.

Bonita, I don't know whether to curse you or kiss you. (Though in fact I
suspect you may be a little young for me. Possibly too young for my kids...)

I haven't written any code just for fun in years. Maybe decades!

I missed Tim's Mod 30 trick, and that might well help. But I think I'm
bored with this. It would save a lot of memory, but the extra
computation might make it slower. Maybe I'll try one day ;)

My code isn't as fast as Bonita's code. Even allowing that I didn't put
any threads in there.

But...

When I compared the output there were differences. Specifically your
program claimed that these numbers

66049,67591,69133,69647,71189,72217,72731,75301,78899,79927,80441,81469,85067,86609,89179,
89693,90721,92263,94319,95861,97403,98431,99973,102029,103057,105113,107683,108197,110767,
111281,112823,113851,115393,117449,118477,118991,120019,123103,125159,126187,128243,129271,
130813,133897,134411,139037,140579,143149,144691,146233,146747,148289,150859,152401,153943,
154457,155999,157541,158569,159083,162167,164737,165251,166279,167821,169363,169877,172961,
173989,175531,177587,180157,182213,184783,186839,188381,189923,190951,193007,194549,195577,
197633,198661,202259,204829,207913,208427,210997,211511,212539,213053,215623,219221,220249,
220763,221791,225389,226417,226931,227959,233099,234127,236183,238753,240809,241837,243379,
244921,248519,249547,251089,252631,254687,256229,259313,260341,261883,262397,264967,265481,
267023,269593,270107,272677,273191,274733,279359,280387,280901,281929,283471,285013,287069,
288611,290153,295807,296321,298891,300947,303517,305059,306601,308657,311741,312769,314311,
315853

are all prime. They aren't. The first one is 257 squared, and the others
all have factors too.

I'd also suggest that you comment demonstration code. It's intended to
show off fancy techniques, and they aren't obvious without explanation.

I learned that lesson when I was a student, when I couldn't understand
something I'd written myself!

Andy

Vir Campestris <vir.campestris@invalid.invalid> writes:
....
> When I compared the output there were differences. Specifically your
> program claimed that these numbers
>
> 66049,67591,69133,69647,71189,72217,72731,75301,78899,79927,80441,81469,85067,86609,89179,
> 89693,90721,92263,94319,95861,97403,98431,99973,102029,103057,105113,107683,108197,110767,
> 111281,112823,113851,115393,117449,118477,118991,120019,123103,125159,126187,128243,129271,
> 130813,133897,134411,139037,140579,143149,144691,146233,146747,148289,150859,152401,153943,
> 154457,155999,157541,158569,159083,162167,164737,165251,166279,167821,169363,169877,172961,
> 173989,175531,177587,180157,182213,184783,186839,188381,189923,190951,193007,194549,195577,
> 197633,198661,202259,204829,207913,208427,210997,211511,212539,213053,215623,219221,220249,
> 220763,221791,225389,226417,226931,227959,233099,234127,236183,238753,240809,241837,243379,
> 244921,248519,249547,251089,252631,254687,256229,259313,260341,261883,262397,264967,265481,
> 267023,269593,270107,272677,273191,274733,279359,280387,280901,281929,283471,285013,287069,
> 288611,290153,295807,296321,298891,300947,303517,305059,306601,308657,311741,312769,314311,
> 315853
>
> are all prime. They aren't. The first one is 257 squared, and the others
> all have factors too.

In fact they are /all/ 257 times some prime. That must be a big clue as
to where the bug is...

--
Ben.