Menu
▾
▴
flint-devel
|
From: Michael A. <mab...@go...> - 2008-12-23 23:36:47
|
Hi Bill, I am just updating to FLINT 1.0.20 in Sage and I noticed an issue I never reported upstream: src/makefile has DOS line endings, i.e. ^M. This causes a BSD make to throw in the towel. One way to fix this via vim would be to to ":s/^M//g: :) subversion should also offer the option to convert all the newlines automatically. Cheers, Michael |
|
From: Michael A. <mab...@go...> - 2008-12-24 00:19:58
|
On Tue, Dec 23, 2008 at 3:36 PM, Michael Abshoff <mab...@go...> wrote: > Hi Bill, > > I am just updating to FLINT 1.0.20 in Sage and I noticed an issue I > never reported upstream: src/makefile has DOS line endings, i.e. ^M. > This causes a BSD make to throw in the towel. > > One way to fix this via vim would be to to ":s/^M//g: :) Ok, I got it all wrong and the problem was a patch I had that caused rejections, so please ignore this dumb mistake. > subversion should also offer the option to convert all the newlines > automatically. > > Cheers, > > Michael I just checked the diff I had for the makefile in Sage vs. vanilla and the only relevant change is this: +libflint.dylib64: $(FLINTOBJ) + $(CC) -m64 -single_module -fPIC -dynamiclib -o libflint.dylib $(FLINTOBJ) $(LIBS) + I.e. I need to add an extra -m64 flag to get a 64 bit OSX dylib. Hopefully we will upgrade to FLINT 1.1 during SD 12, but we will see how that goes. Cheers, Michael |
|
From: William H. <ha...@ya...> - 2008-12-24 02:32:09
|
Thanks for the report. I've merged this into FLINT 1.1.0 and FLINT 1.2-devel. As no one else is likely to have hit this problem yet and we just released a major version, I did not update the minor revision number. But I did update the CHANGELOG, effectively tacking the fix on to the end of the major revision. Bill. --- On Wed, 12/24/08, Michael Abshoff <mab...@go...> wrote: > From: Michael Abshoff <mab...@go...> > Subject: Re: [flint-devel] newlines in makefile > To: "Development list for FLINT" <fli...@li...> > Date: Wednesday, December 24, 2008, 11:19 AM > On Tue, Dec 23, 2008 at 3:36 PM, Michael Abshoff > <mab...@go...> wrote: > > Hi Bill, > > > > I am just updating to FLINT 1.0.20 in Sage and I > noticed an issue I > > never reported upstream: src/makefile has DOS line > endings, i.e. ^M. > > This causes a BSD make to throw in the towel. > > > > One way to fix this via vim would be to to > ":s/^M//g: :) > > Ok, I got it all wrong and the problem was a patch I had > that caused > rejections, so please ignore this dumb mistake. > > > subversion should also offer the option to convert all > the newlines > > automatically. > > > > Cheers, > > > > Michael > > I just checked the diff I had for the makefile in Sage vs. > vanilla and > the only relevant change is this: > > +libflint.dylib64: $(FLINTOBJ) > + $(CC) -m64 -single_module -fPIC -dynamiclib -o > libflint.dylib > $(FLINTOBJ) $(LIBS) > + > > I.e. I need to add an extra -m64 flag to get a 64 bit OSX > dylib. > > Hopefully we will upgrade to FLINT 1.1 during SD 12, but we > will see > how that goes. > > Cheers, > > Michael > > ------------------------------------------------------------------------------ > _______________________________________________ > flint-devel mailing list > fli...@li... > https://lists.sourceforge.net/lists/listinfo/flint-devel |
|
From: Burcin E. <bu...@er...> - 2008-12-25 13:10:28
|
On Tue, 23 Dec 2008 16:19:53 -0800 "Michael Abshoff" <mab...@go...> wrote: > On Tue, Dec 23, 2008 at 3:36 PM, Michael Abshoff > <mab...@go...> wrote: <snip> > Hopefully we will upgrade to FLINT 1.1 during SD 12, but we will see > how that goes. Do you have an spkg for 1.1 already? If you send it my way I can quickly see if the update breaks anything, and move ZZ[x] polynomials to use FLINT for factoring as well. I will also wait for the updated spkg to push my zmod_poly wrapper upstream. The new derivative and factor methods will save me some time. Cheers, Burcin |
|
From: Michael A. <mab...@go...> - 2008-12-25 13:28:26
|
On Thu, Dec 25, 2008 at 4:53 AM, Burcin Erocal <bu...@er...> wrote: > On Tue, 23 Dec 2008 16:19:53 -0800 > "Michael Abshoff" <mab...@go...> wrote: Hi Burcin, >> On Tue, Dec 23, 2008 at 3:36 PM, Michael Abshoff >> <mab...@go...> wrote: > <snip> >> Hopefully we will upgrade to FLINT 1.1 during SD 12, but we will see >> how that goes. > > Do you have an spkg for 1.1 already? Nope, but it shouldn't be too hard to do. We need to fix some dependency issues due to David's zn_poly, but other than that I wanted to clean up spkg-install and spkg-check in Sage's flint.spkg since they are pretty messy. > If you send it my way I can > quickly see if the update breaks anything, and move ZZ[x] polynomials > to use FLINT for factoring as well. Cool. > I will also wait for the updated spkg to push my zmod_poly wrapper > upstream. The new derivative and factor methods will save me some time. :) > Cheers, > > Burcin Cheers, Michael > ------------------------------------------------------------------------------ > _______________________________________________ > flint-devel mailing list > fli...@li... > https://lists.sourceforge.net/lists/listinfo/flint-devel > |
|
From: Bill H. <goo...@go...> - 2008-12-25 14:39:47
|
There's only Z/pZ[x] factoring so far, not Z[x]. I'll fix the test failures as soon as I can. What architecture was it? I *think* the failure is probably due to using 32 bit primes not 31 bit, but I need to check it out. Bill. On 25/12/2008, Michael Abshoff <mab...@go...> wrote: > On Thu, Dec 25, 2008 at 4:53 AM, Burcin Erocal <bu...@er...> wrote: >> On Tue, 23 Dec 2008 16:19:53 -0800 >> "Michael Abshoff" <mab...@go...> wrote: > > Hi Burcin, > >>> On Tue, Dec 23, 2008 at 3:36 PM, Michael Abshoff >>> <mab...@go...> wrote: >> <snip> >>> Hopefully we will upgrade to FLINT 1.1 during SD 12, but we will see >>> how that goes. >> >> Do you have an spkg for 1.1 already? > > Nope, but it shouldn't be too hard to do. We need to fix some > dependency issues due to David's zn_poly, but other than that I wanted > to clean up spkg-install and spkg-check in Sage's flint.spkg since > they are pretty messy. > >> If you send it my way I can >> quickly see if the update breaks anything, and move ZZ[x] polynomials >> to use FLINT for factoring as well. > > Cool. > >> I will also wait for the updated spkg to push my zmod_poly wrapper >> upstream. The new derivative and factor methods will save me some time. > > :) > >> Cheers, >> >> Burcin > > Cheers, > > Michael > >> ------------------------------------------------------------------------------ >> _______________________________________________ >> flint-devel mailing list >> fli...@li... >> https://lists.sourceforge.net/lists/listinfo/flint-devel >> > > ------------------------------------------------------------------------------ > _______________________________________________ > flint-devel mailing list > fli...@li... > https://lists.sourceforge.net/lists/listinfo/flint-devel > |
|
From: Burcin E. <bu...@er...> - 2008-12-25 14:51:01
|
On Thu, 25 Dec 2008 14:39:37 +0000 "Bill Hart" <goo...@go...> wrote: > There's only Z/pZ[x] factoring so far, not Z[x]. I'll fix the test > failures as soon as I can. What architecture was it? I *think* the > failure is probably due to using 32 bit primes not 31 bit, but I need > to check it out. OK, good to know. I only saw the factor functions when I was skimming the zmod_poly section for the Zn[x] wrapper in Sage, and somehow assumed they were available for Z[x] as well. This means even less work to update to 1.1.0 instead of 1.0.20. :) FWIW, I tested 1.1.0 on some of the 32-bit machines I have access to, and all tests passed. Cheers, Burcin |
|
From: Michael A. <mab...@go...> - 2008-12-25 15:47:09
|
On Thu, Dec 25, 2008 at 6:50 AM, Burcin Erocal <bu...@er...> wrote: > On Thu, 25 Dec 2008 14:39:37 +0000 > "Bill Hart" <goo...@go...> wrote: Hi, >> There's only Z/pZ[x] factoring so far, not Z[x]. I'll fix the test >> failures as soon as I can. What architecture was it? I *think* the >> failure is probably due to using 32 bit primes not 31 bit, but I need >> to check it out. Ok. > OK, good to know. I only saw the factor functions when I was skimming > the zmod_poly section for the Zn[x] wrapper in Sage, and somehow > assumed they were available for Z[x] as well. > > This means even less work to update to 1.1.0 instead of 1.0.20. :) > FWIW, I tested 1.1.0 on some of the 32-bit machines I have access to, > and all tests passed. Sure, but I am reluctant to switch to 1.1 so late in the game for 3.2.3 given the need to do 3.3 in less than a week. So if the 1.1 update is more or less painless I would want to do that in 3.3, but if anything is turned up by valgrinding 1.0.20 on 32 bit boxen I would be curious if this is worth fixing via 1.0.21 in 3.2.3 or if we should just move to 1.1 in 3.3. > > Cheers, > > Burcin Cheers, Michael > > ------------------------------------------------------------------------------ > _______________________________________________ > flint-devel mailing list > fli...@li... > https://lists.sourceforge.net/lists/listinfo/flint-devel > |
|
From: Bill H. <goo...@go...> - 2008-12-25 16:36:26
|
I'll reply to both posts at once. The test failures will be cases hit because of different random seeds on those 32 machines. They should be trivial to fix. There is an outside chance I can do it tonight, but if I have access to the machines involved, this will be easier. Upgrading to 1.1 should not be painful, unless you want to incorporate lots of the new functionality (which I hope you will :-)). In that case there's the wrapper to write. But otherwise things should be quite stable. I've done very long runs with most of the new functions, and no problems. I don't think SAGE uses any of the functionality where the interface has changed. I've only just noticed that these failures are being reported for FLINT 1.0.20. That's odd, because only bug fixes have gone into that branch for some time now, so presumably these bugs have been there for a while, and have been triggered due to different random seeds after rebuilding. If the bugs are in 1.0.20 they are also likely in 1.1, just not triggered because of different random seeds. Bill. 2008/12/25 Michael Abshoff <mab...@go...>: > On Thu, Dec 25, 2008 at 6:50 AM, Burcin Erocal <bu...@er...> wrote: >> On Thu, 25 Dec 2008 14:39:37 +0000 >> "Bill Hart" <goo...@go...> wrote: > > Hi, > >>> There's only Z/pZ[x] factoring so far, not Z[x]. I'll fix the test >>> failures as soon as I can. What architecture was it? I *think* the >>> failure is probably due to using 32 bit primes not 31 bit, but I need >>> to check it out. > > Ok. > >> OK, good to know. I only saw the factor functions when I was skimming >> the zmod_poly section for the Zn[x] wrapper in Sage, and somehow >> assumed they were available for Z[x] as well. >> >> This means even less work to update to 1.1.0 instead of 1.0.20. :) >> FWIW, I tested 1.1.0 on some of the 32-bit machines I have access to, >> and all tests passed. > > Sure, but I am reluctant to switch to 1.1 so late in the game for > 3.2.3 given the need to do 3.3 in less than a week. So if the 1.1 > update is more or less painless I would want to do that in 3.3, but if > anything is turned up by valgrinding 1.0.20 on 32 bit boxen I would be > curious if this is worth fixing via 1.0.21 in 3.2.3 or if we should > just move to 1.1 in 3.3. > >> >> Cheers, >> >> Burcin > > Cheers, > > Michael > >> >> ------------------------------------------------------------------------------ >> _______________________________________________ >> flint-devel mailing list >> fli...@li... >> https://lists.sourceforge.net/lists/listinfo/flint-devel >> > > ------------------------------------------------------------------------------ > _______________________________________________ > flint-devel mailing list > fli...@li... > https://lists.sourceforge.net/lists/listinfo/flint-devel > |
|
From: Bill H. <goo...@go...> - 2008-12-25 16:45:05
|
OK, I see I do have access. Didn't read the posts properly on my IPhone. Back home now. Bill. 2008/12/25 Bill Hart <goo...@go...>: > I'll reply to both posts at once. > > The test failures will be cases hit because of different random seeds > on those 32 machines. They should be trivial to fix. There is an > outside chance I can do it tonight, but if I have access to the > machines involved, this will be easier. > > Upgrading to 1.1 should not be painful, unless you want to incorporate > lots of the new functionality (which I hope you will :-)). In that > case there's the wrapper to write. But otherwise things should be > quite stable. I've done very long runs with most of the new functions, > and no problems. I don't think SAGE uses any of the functionality > where the interface has changed. > > I've only just noticed that these failures are being reported for > FLINT 1.0.20. That's odd, because only bug fixes have gone into that > branch for some time now, so presumably these bugs have been there for > a while, and have been triggered due to different random seeds after > rebuilding. > > If the bugs are in 1.0.20 they are also likely in 1.1, just not > triggered because of different random seeds. > > Bill. > > 2008/12/25 Michael Abshoff <mab...@go...>: >> On Thu, Dec 25, 2008 at 6:50 AM, Burcin Erocal <bu...@er...> wrote: >>> On Thu, 25 Dec 2008 14:39:37 +0000 >>> "Bill Hart" <goo...@go...> wrote: >> >> Hi, >> >>>> There's only Z/pZ[x] factoring so far, not Z[x]. I'll fix the test >>>> failures as soon as I can. What architecture was it? I *think* the >>>> failure is probably due to using 32 bit primes not 31 bit, but I need >>>> to check it out. >> >> Ok. >> >>> OK, good to know. I only saw the factor functions when I was skimming >>> the zmod_poly section for the Zn[x] wrapper in Sage, and somehow >>> assumed they were available for Z[x] as well. >>> >>> This means even less work to update to 1.1.0 instead of 1.0.20. :) >>> FWIW, I tested 1.1.0 on some of the 32-bit machines I have access to, >>> and all tests passed. >> >> Sure, but I am reluctant to switch to 1.1 so late in the game for >> 3.2.3 given the need to do 3.3 in less than a week. So if the 1.1 >> update is more or less painless I would want to do that in 3.3, but if >> anything is turned up by valgrinding 1.0.20 on 32 bit boxen I would be >> curious if this is worth fixing via 1.0.21 in 3.2.3 or if we should >> just move to 1.1 in 3.3. >> >>> >>> Cheers, >>> >>> Burcin >> >> Cheers, >> >> Michael >> >>> >>> ------------------------------------------------------------------------------ >>> _______________________________________________ >>> flint-devel mailing list >>> fli...@li... >>> https://lists.sourceforge.net/lists/listinfo/flint-devel >>> >> >> ------------------------------------------------------------------------------ >> _______________________________________________ >> flint-devel mailing list >> fli...@li... >> https://lists.sourceforge.net/lists/listinfo/flint-devel >> > |
|
From: Michael A. <mab...@go...> - 2008-12-25 17:27:02
|
On Thu, Dec 25, 2008 at 8:36 AM, Bill Hart <goo...@go...> wrote: Hi Bill, > I'll reply to both posts at once. Ok. > The test failures will be cases hit because of different random seeds > on those 32 machines. They should be trivial to fix. There is an > outside chance I can do it tonight, but if I have access to the > machines involved, this will be easier. The linux box is cicero on SkyNet, but I am currently valgrind there, so I am keeping that box busy. The other one is varro also on SkyNet, but I don't think your account works there, so you should ping Mariah. > Upgrading to 1.1 should not be painful, unless you want to incorporate > lots of the new functionality (which I hope you will :-)). In that > case there's the wrapper to write. But otherwise things should be > quite stable. I've done very long runs with most of the new functions, > and no problems. I don't think SAGE uses any of the functionality > where the interface has changed. Ok. > I've only just noticed that these failures are being reported for > FLINT 1.0.20. That's odd, because only bug fixes have gone into that > branch for some time now, so presumably these bugs have been there for > a while, and have been triggered due to different random seeds after > rebuilding. Yes, I just turned tests on since I figured it cannot hurt and I was just as surprised that on both 32 bit test boxen things didn't work too well. > If the bugs are in 1.0.20 they are also likely in 1.1, just not > triggered because of different random seeds. Good to know. Once I am done with 1.0.20 I will test 1.1 on the same box. > Bill. Cheers, Michael |
|
From: Bill H. <goo...@go...> - 2008-12-25 17:44:07
|
I've fixed the bug (the same bug caused both problems) and will issue FLINT 1.0.21 in a couple of minutes. It was as I thought, using 32 bit primes instead of 31 bits. This bug does not exist in FLINT 1.1 or 1.2. Bill. 2008/12/25 Michael Abshoff <mab...@go...>: > On Thu, Dec 25, 2008 at 8:36 AM, Bill Hart <goo...@go...> wrote: > > Hi Bill, > >> I'll reply to both posts at once. > > Ok. > >> The test failures will be cases hit because of different random seeds >> on those 32 machines. They should be trivial to fix. There is an >> outside chance I can do it tonight, but if I have access to the >> machines involved, this will be easier. > > The linux box is cicero on SkyNet, but I am currently valgrind there, > so I am keeping that box busy. The other one is varro also on SkyNet, > but I don't think your account works there, so you should ping Mariah. > >> Upgrading to 1.1 should not be painful, unless you want to incorporate >> lots of the new functionality (which I hope you will :-)). In that >> case there's the wrapper to write. But otherwise things should be >> quite stable. I've done very long runs with most of the new functions, >> and no problems. I don't think SAGE uses any of the functionality >> where the interface has changed. > > Ok. > >> I've only just noticed that these failures are being reported for >> FLINT 1.0.20. That's odd, because only bug fixes have gone into that >> branch for some time now, so presumably these bugs have been there for >> a while, and have been triggered due to different random seeds after >> rebuilding. > > Yes, I just turned tests on since I figured it cannot hurt and I was > just as surprised that on both 32 bit test boxen things didn't work > too well. > >> If the bugs are in 1.0.20 they are also likely in 1.1, just not >> triggered because of different random seeds. > > Good to know. Once I am done with 1.0.20 I will test 1.1 on the same box. > >> Bill. > > Cheers, > > Michael > > ------------------------------------------------------------------------------ > _______________________________________________ > flint-devel mailing list > fli...@li... > https://lists.sourceforge.net/lists/listinfo/flint-devel > |
|
From: Bill H. <goo...@go...> - 2008-12-25 17:54:15
|
OK, I've issued 1.0.21 which fixes these issues. It isn't immediately clear to me whether this bug was in previous versions or not. It wasn't an introduced bug as such, but it may have been created when another bug was fixed. Either way, things are done differently in later versions of FLINT and so they won't be an issue. Bill. 2008/12/25 Bill Hart <goo...@go...>: > I've fixed the bug (the same bug caused both problems) and will issue > FLINT 1.0.21 in a couple of minutes. It was as I thought, using 32 bit > primes instead of 31 bits. > > This bug does not exist in FLINT 1.1 or 1.2. > > Bill. > > 2008/12/25 Michael Abshoff <mab...@go...>: >> On Thu, Dec 25, 2008 at 8:36 AM, Bill Hart <goo...@go...> wrote: >> >> Hi Bill, >> >>> I'll reply to both posts at once. >> >> Ok. >> >>> The test failures will be cases hit because of different random seeds >>> on those 32 machines. They should be trivial to fix. There is an >>> outside chance I can do it tonight, but if I have access to the >>> machines involved, this will be easier. >> >> The linux box is cicero on SkyNet, but I am currently valgrind there, >> so I am keeping that box busy. The other one is varro also on SkyNet, >> but I don't think your account works there, so you should ping Mariah. >> >>> Upgrading to 1.1 should not be painful, unless you want to incorporate >>> lots of the new functionality (which I hope you will :-)). In that >>> case there's the wrapper to write. But otherwise things should be >>> quite stable. I've done very long runs with most of the new functions, >>> and no problems. I don't think SAGE uses any of the functionality >>> where the interface has changed. >> >> Ok. >> >>> I've only just noticed that these failures are being reported for >>> FLINT 1.0.20. That's odd, because only bug fixes have gone into that >>> branch for some time now, so presumably these bugs have been there for >>> a while, and have been triggered due to different random seeds after >>> rebuilding. >> >> Yes, I just turned tests on since I figured it cannot hurt and I was >> just as surprised that on both 32 bit test boxen things didn't work >> too well. >> >>> If the bugs are in 1.0.20 they are also likely in 1.1, just not >>> triggered because of different random seeds. >> >> Good to know. Once I am done with 1.0.20 I will test 1.1 on the same box. >> >>> Bill. >> >> Cheers, >> >> Michael >> >> ------------------------------------------------------------------------------ >> _______________________________________________ >> flint-devel mailing list >> fli...@li... >> https://lists.sourceforge.net/lists/listinfo/flint-devel >> > |
|
From: Michael A. <mab...@go...> - 2008-12-25 18:04:04
|
On Thu, Dec 25, 2008 at 9:54 AM, Bill Hart <goo...@go...> wrote: > OK, I've issued 1.0.21 which fixes these issues. It isn't immediately > clear to me whether this bug was in previous versions or not. It > wasn't an introduced bug as such, but it may have been created when > another bug was fixed. Either way, things are done differently in > later versions of FLINT and so they won't be an issue. > > Bill. Thanks, I will update the spkg and try again. Cheers, Michael |
Thanks for helping keep SourceForge clean.
X