<div dir="ltr"><div>Yes, sorry if I was not clear, but you got it :)</div><div><br></div><div>Thank you,</div><div>Florin<br></div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Tue, Jan 11, 2022 at 7:43 AM Marcel Taeumel <<a href="mailto:marcel.taeumel@hpi.de">marcel.taeumel@hpi.de</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"> <div><div id="gmail-m_-3863012451645651506__MailbirdStyleContent" style="font-size:10pt;font-family:Arial;color:rgb(0,0,0);text-align:left" dir="ltr">
Hi Florin --<div><br></div><div>Not sure whether you suggested a fix but ... I just have to replace "readStreamInstSize" with "writeStreamInstSize", too. I will do that.</div><div><br></div><div>Best,</div><div>Marcel</div><div></div><blockquote type="cite" style="border-left-style:solid;border-width:1px;margin-top:20px;margin-left:0px;padding-left:10px">
<p style="color:rgb(170,170,170);margin-top:10px">Am 11.01.2022 13:27:50 schrieb Florin Mateoc <<a href="mailto:florin.mateoc@gmail.com" target="_blank">florin.mateoc@gmail.com</a>>:</p><div style="font-family:Arial,Helvetica,sans-serif"> Hi Marcel,
<br>
<br>Thank you for the quick response, but I am afraid the fix in
<br>#primitiveZipSendBlock is still wrong.
<br>While the method is working with multiple types of objects, in the rcvr
<br>variable it clearly loads a zip encoder, and only a zip encoder:
<br>
<br> (self loadZipEncoderFrom: rcvr)
<br>
<br>And at the end of the method, it is in this ZipEncoder instance rcvr that
<br>it is storing the modified values.
<br>ZipEncoder is a subclass of WriteStream, not of ReadStream.
<br>
<br>All the best,
<br>Florin
<br>
<br>On Tue, Jan 11, 2022 at 6:22 AM Marcel Taeumel <u></u>
<br>wrote:
<br>
<br>>
<br>> Hi Florin, hi Fabio --
<br>>
<br>> Fixed via VMMaker.oscog-mt.3135.
<br>>
<br>> Thanks for reporting this!
<br>>
<br>> Best,
<br>> Marcel
<br>>
<br>> Am 10.01.2022 19:39:46 schrieb Florin Mateoc <u></u>:
<br>> Also now MCSerializationTest>>#testMczSerialization is succeeding for me,
<br>> it was failing before (in JsSqueak)
<br>>
<br>> On Mon, Jan 10, 2022 at 3:42 AM Fabio Niephaus wrote:
<br>>
<br>> >
<br>> > Hi Florin,
<br>> >
<br>> > Happy new year and thanks for the bug report! PNGReadWriterTest seems
<br>> > to pass on TruffleSqueak (although I remember some bug fixing a while
<br>> > ago). Anyway, I looked into #primitiveZipSendBlock which seems to be
<br>> > the only sender of #loadZipEncoderFrom:.
<br>> >
<br>> > While #loadZipEncoderFrom: reads from writeStreamInstSize + 1 and + 2,
<br>> > which is off by one as you mentioned, #primitiveZipSendBlock stores at
<br>> > readStreamInstSize + 1 and + 2. And as it happens, readStreamInstSize
<br>> > seems to be 4 while writeStreamInstSize is 5, which compensates for
<br>> > the off-by-one error. However, I don't understand why reading from the
<br>> > wrong positions does not cause the tests to fail. Anyway, I cleaned
<br>> > this up in TruffleSqueak (see [1]).
<br>> >
<br>> > Fabio
<br>> >
<br>> > [1]
<br>> >
<br>> <a href="https://github.com/hpi-swa/trufflesqueak/commit/5b6066b54e03e19ec4c4d2b22550ff081e46f6e6" target="_blank">https://github.com/hpi-swa/trufflesqueak/commit/5b6066b54e03e19ec4c4d2b22550ff081e46f6e6</a>
<br>> >
<br>> > On Mon, Jan 10, 2022 at 4:33 AM Florin Mateoc
<br>> > wrote:
<br>> > >
<br>> > >
<br>> > > I should have also mentioned that PNGReadWriterTest also has 18
<br>> failures
<br>> > for SqueakJS, and I think I remember them failing in TruffleSqueak as
<br>> well
<br>> > >
<br>> > > On Sun, Jan 9, 2022 at 10:18 PM Florin Mateoc
<br>> > wrote:
<br>> > >>
<br>> > >> Hi all, and Happy New Year!
<br>> > >>
<br>> > >> I think I just found a couple of bugs in the DeflatePlugin. They were
<br>> > causing errors in my JsSqueak, I don't fully understand why they don't
<br>> seem
<br>> > to cause problems in Squeak. Anyway, after I fixed them, the primitives
<br>> > work and the PNGReadWriterTest tests are all succeeding for me.
<br>> > >>
<br>> > >> One is in #loadZipEncoderFrom:
<br>> > >> The method checks and then determines the writeStreamInstSize and
<br>> then
<br>> > fetches the bitBuffer and bitPosition (the first two instvars of
<br>> > ZipEncoder, a direct subclass of WriteStream) from offsets
<br>> > writeStreamInstSize + 1 and writeStreamInstSize + 2, although the
<br>> fetches
<br>> > are 0-based
<br>> > >>
<br>> > >> The other is in #primitiveZipSendBlock
<br>> > >> The method, still working on a ZipEncoder, at the end stores the
<br>> > modified values for bitBuffer and bitPosition at offset
<br>> readStreamInstSize
<br>> > + 1 and readStreamInstSize + 2, which is just wrong.
<br>> > >>
<br>> > >> As I said, I don't quite get why this does not cause at least
<br>> > image-side errors if not VM crashes in Squeak, but they should probably
<br>> be
<br>> > fixed anyway :)
<br>> > >>
<br>> > >> All the best,
<br>> > >> Florin
<br>> >
<br>> Also now MCSerializationTest>>#testMczSerialization is succeeding for me,
<br>> it was failing before
<br>> (in JsSqueak)
<br>>
<br>>
<br>> On Mon, Jan 10, 2022 at 3:42 AM Fabio Niephaus <u></u>
<br>> wrote:
<br>>
<br>>>
<br>>>
<br>>> Hi Florin,
<br>>>
<br>>>
<br>>>
<br>>> Happy new year and thanks for the bug report! PNGReadWriterTest seems
<br>>>
<br>>> to pass on TruffleSqueak (although I remember some bug fixing a while
<br>>>
<br>>> ago). Anyway, I looked into #primitiveZipSendBlock which seems to be
<br>>>
<br>>> the only sender of #loadZipEncoderFrom:.
<br>>>
<br>>>
<br>>>
<br>>> While #loadZipEncoderFrom: reads from writeStreamInstSize + 1 and + 2,
<br>>>
<br>>> which is off by one as you mentioned, #primitiveZipSendBlock stores at
<br>>>
<br>>> readStreamInstSize + 1 and + 2. And as it happens, readStreamInstSize
<br>>>
<br>>> seems to be 4 while writeStreamInstSize is 5, which compensates for
<br>>>
<br>>> the off-by-one error. However, I don't understand why reading from the
<br>>>
<br>>> wrong positions does not cause the tests to fail. Anyway, I cleaned
<br>>>
<br>>> this up in TruffleSqueak (see [1]).
<br>>>
<br>>>
<br>>>
<br>>> Fabio
<br>>>
<br>>>
<br>>>
<br>>> [1]
<br>>> <a href="https://github.com/hpi-swa/trufflesqueak/commit/5b6066b54e03e19ec4c4d2b22550ff081e46f6e6" target="_blank">https://github.com/hpi-swa/trufflesqueak/commit/5b6066b54e03e19ec4c4d2b22550ff081e46f6e6</a>
<br>>>
<br>>>
<br>>>
<br>>> On Mon, Jan 10, 2022 at 4:33 AM Florin Mateoc <u></u>
<br>>> wrote:
<br>>>
<br>>> >
<br>>>
<br>>> >
<br>>>
<br>>> > I should have also mentioned that PNGReadWriterTest also has 18
<br>>> failures for SqueakJS, and I think I remember them failing in TruffleSqueak
<br>>> as well
<br>>>
<br>>> >
<br>>>
<br>>> > On Sun, Jan 9, 2022 at 10:18 PM Florin Mateoc <u></u>
<br>>> wrote:
<br>>>
<br>>> >>
<br>>>
<br>>> >> Hi all, and Happy New Year!
<br>>>
<br>>> >>
<br>>>
<br>>> >> I think I just found a couple of bugs in the DeflatePlugin. They were
<br>>> causing errors in my JsSqueak, I don't fully understand why they don't seem
<br>>> to cause problems in Squeak. Anyway, after I fixed them, the primitives
<br>>> work and the PNGReadWriterTest tests are all succeeding for me.
<br>>>
<br>>> >>
<br>>>
<br>>> >> One is in #loadZipEncoderFrom:
<br>>>
<br>>> >> The method checks and then determines the writeStreamInstSize and then
<br>>> fetches the bitBuffer and bitPosition (the first two instvars of
<br>>> ZipEncoder, a direct subclass of WriteStream) from offsets
<br>>> writeStreamInstSize + 1 and writeStreamInstSize + 2, although the fetches
<br>>> are 0-based
<br>>>
<br>>> >>
<br>>>
<br>>> >> The other is in #primitiveZipSendBlock
<br>>>
<br>>> >> The method, still working on a ZipEncoder, at the end stores the
<br>>> modified values for bitBuffer and bitPosition at offset readStreamInstSize
<br>>> + 1 and readStreamInstSize + 2, which is just wrong.
<br>>>
<br>>> >>
<br>>>
<br>>> >> As I said, I don't quite get why this does not cause at least
<br>>> image-side errors if not VM crashes in Squeak, but they should probably be
<br>>> fixed anyway :)
<br>>>
<br>>> >>
<br>>>
<br>>> >> All the best,
<br>>>
<br>>> >> Florin
<br>>>
<br>>>
<br>>
<br><div dir="ltr"><div>Hi Marcel,</div><div><br></div><div>Thank you for the quick response, but I am afraid the fix in
<br><span>#primitiveZipSendBlock</span> is still wrong.</div><div>While the method is working with multiple types of objects, in the rcvr variable it clearly loads a zip encoder, and only a zip encoder:</div><div></div><div>
<br>
<br><pre style="white-space:pre-wrap;color:rgb(0,0,0);font-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;font-weight:400;letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;word-spacing:0px;text-decoration-style:initial;text-decoration-color:initial"> (self loadZipEncoderFrom: rcvr)
<br></pre>And at the end of the method, it is in this ZipEncoder instance rcvr that it is storing the modified values.</div><div>ZipEncoder is a subclass of WriteStream, not of ReadStream.</div><div><br></div><div>All the best,</div><div>Florin<br></div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Tue, Jan 11, 2022 at 6:22 AM Marcel Taeumel <<a href="mailto:marcel.taeumel@hpi.de" target="_blank">marcel.taeumel@hpi.de</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"> <div id="gmail-m_-3863012451645651506gmail-m_-7492184690558884719__MailbirdStyleContent" style="font-size:10pt;font-family:Arial;color:rgb(0,0,0);text-align:left" dir="ltr">
<br> Hi Florin, hi Fabio --<div><br></div><div>Fixed via VMMaker.oscog-mt.3135.</div><div><br></div><div>Thanks for reporting this!</div><div><br></div><div>Best,</div><div>Marcel</div><div></div>
<br> <blockquote type="cite" style="border-left-style:solid;border-width:1px;margin-top:20px;margin-left:0px;padding-left:10px;min-width:500px">
<br> <p style="color:rgb(170,170,170);margin-top:10px">Am 10.01.2022 19:39:46 schrieb Florin Mateoc <<a href="mailto:florin.mateoc@gmail.com" target="_blank">florin.mateoc@gmail.com</a>>:</p><div style="font-family:Arial,Helvetica,sans-serif"> Also now MCSerializationTest>>#testMczSerialization is succeeding for me,
<br><br>it was failing before (in JsSqueak)
<br><br>
<br><br>On Mon, Jan 10, 2022 at 3:42 AM Fabio Niephaus <u></u> wrote:
<br><br>
<br><br>>
<br><br>> Hi Florin,
<br><br>>
<br><br>> Happy new year and thanks for the bug report! PNGReadWriterTest seems
<br><br>> to pass on TruffleSqueak (although I remember some bug fixing a while
<br><br>> ago). Anyway, I looked into #primitiveZipSendBlock which seems to be
<br><br>> the only sender of #loadZipEncoderFrom:.
<br><br>>
<br><br>> While #loadZipEncoderFrom: reads from writeStreamInstSize + 1 and + 2,
<br><br>> which is off by one as you mentioned, #primitiveZipSendBlock stores at
<br><br>> readStreamInstSize + 1 and + 2. And as it happens, readStreamInstSize
<br><br>> seems to be 4 while writeStreamInstSize is 5, which compensates for
<br><br>> the off-by-one error. However, I don't understand why reading from the
<br><br>> wrong positions does not cause the tests to fail. Anyway, I cleaned
<br><br>> this up in TruffleSqueak (see [1]).
<br><br>>
<br><br>> Fabio
<br><br>>
<br><br>> [1]
<br><br>> <a href="https://github.com/hpi-swa/trufflesqueak/commit/5b6066b54e03e19ec4c4d2b22550ff081e46f6e6" target="_blank">https://github.com/hpi-swa/trufflesqueak/commit/5b6066b54e03e19ec4c4d2b22550ff081e46f6e6</a>
<br><br>>
<br><br>> On Mon, Jan 10, 2022 at 4:33 AM Florin Mateoc <u></u>
<br><br>> wrote:
<br><br>> >
<br><br>> >
<br><br>> > I should have also mentioned that PNGReadWriterTest also has 18 failures
<br><br>> for SqueakJS, and I think I remember them failing in TruffleSqueak as well
<br><br>> >
<br><br>> > On Sun, Jan 9, 2022 at 10:18 PM Florin Mateoc <u></u>
<br><br>> wrote:
<br><br>> >>
<br><br>> >> Hi all, and Happy New Year!
<br><br>> >>
<br><br>> >> I think I just found a couple of bugs in the DeflatePlugin. They were
<br><br>> causing errors in my JsSqueak, I don't fully understand why they don't seem
<br><br>> to cause problems in Squeak. Anyway, after I fixed them, the primitives
<br><br>> work and the PNGReadWriterTest tests are all succeeding for me.
<br><br>> >>
<br><br>> >> One is in #loadZipEncoderFrom:
<br><br>> >> The method checks and then determines the writeStreamInstSize and then
<br><br>> fetches the bitBuffer and bitPosition (the first two instvars of
<br><br>> ZipEncoder, a direct subclass of WriteStream) from offsets
<br><br>> writeStreamInstSize + 1 and writeStreamInstSize + 2, although the fetches
<br><br>> are 0-based
<br><br>> >>
<br><br>> >> The other is in #primitiveZipSendBlock
<br><br>> >> The method, still working on a ZipEncoder, at the end stores the
<br><br>> modified values for bitBuffer and bitPosition at offset readStreamInstSize
<br><br>> + 1 and readStreamInstSize + 2, which is just wrong.
<br><br>> >>
<br><br>> >> As I said, I don't quite get why this does not cause at least
<br><br>> image-side errors if not VM crashes in Squeak, but they should probably be
<br><br>> fixed anyway :)
<br><br>> >>
<br><br>> >> All the best,
<br><br>> >> Florin
<br><br>>
<br><br><div dir="ltr">Also now MCSerializationTest>>#testMczSerialization is succeeding for me, it was failing before
<br><br> (in JsSqueak)
<br><br>
<br><br></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Mon, Jan 10, 2022 at 3:42 AM Fabio Niephaus <<a href="mailto:lists@fniephaus.com" target="_blank">lists@fniephaus.com</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex;min-width:500px"> <br>
<br><br>Hi Florin,<br>
<br><br><br>
<br><br>Happy new year and thanks for the bug report! PNGReadWriterTest seems<br>
<br><br>to pass on TruffleSqueak (although I remember some bug fixing a while<br>
<br><br>ago). Anyway, I looked into #primitiveZipSendBlock which seems to be<br>
<br><br>the only sender of #loadZipEncoderFrom:.<br>
<br><br><br>
<br><br>While #loadZipEncoderFrom: reads from writeStreamInstSize + 1 and + 2,<br>
<br><br>which is off by one as you mentioned, #primitiveZipSendBlock stores at<br>
<br><br>readStreamInstSize + 1 and + 2. And as it happens, readStreamInstSize<br>
<br><br>seems to be 4 while writeStreamInstSize is 5, which compensates for<br>
<br><br>the off-by-one error. However, I don't understand why reading from the<br>
<br><br>wrong positions does not cause the tests to fail. Anyway, I cleaned<br>
<br><br>this up in TruffleSqueak (see [1]).<br>
<br><br><br>
<br><br>Fabio<br>
<br><br><br>
<br><br>[1] <a href="https://github.com/hpi-swa/trufflesqueak/commit/5b6066b54e03e19ec4c4d2b22550ff081e46f6e6" rel="noreferrer" target="_blank">https://github.com/hpi-swa/trufflesqueak/commit/5b6066b54e03e19ec4c4d2b22550ff081e46f6e6</a><br>
<br><br><br>
<br><br>On Mon, Jan 10, 2022 at 4:33 AM Florin Mateoc <<a href="mailto:florin.mateoc@gmail.com" target="_blank">florin.mateoc@gmail.com</a>> wrote:<br>
<br><br>><br>
<br><br>><br>
<br><br>> I should have also mentioned that PNGReadWriterTest also has 18 failures for SqueakJS, and I think I remember them failing in TruffleSqueak as well<br>
<br><br>><br>
<br><br>> On Sun, Jan 9, 2022 at 10:18 PM Florin Mateoc <<a href="mailto:florin.mateoc@gmail.com" target="_blank">florin.mateoc@gmail.com</a>> wrote:<br>
<br><br>>><br>
<br><br>>> Hi all, and Happy New Year!<br>
<br><br>>><br>
<br><br>>> I think I just found a couple of bugs in the DeflatePlugin. They were causing errors in my JsSqueak, I don't fully understand why they don't seem to cause problems in Squeak. Anyway, after I fixed them, the primitives work and the PNGReadWriterTest tests are all succeeding for me.<br>
<br><br>>><br>
<br><br>>> One is in #loadZipEncoderFrom:<br>
<br><br>>> The method checks and then determines the writeStreamInstSize and then fetches the bitBuffer and bitPosition (the first two instvars of ZipEncoder, a direct subclass of WriteStream) from offsets writeStreamInstSize + 1 and writeStreamInstSize + 2, although the fetches are 0-based<br>
<br><br>>><br>
<br><br>>> The other is in #primitiveZipSendBlock<br>
<br><br>>> The method, still working on a ZipEncoder, at the end stores the modified values for bitBuffer and bitPosition at offset readStreamInstSize + 1 and readStreamInstSize + 2, which is just wrong.<br>
<br><br>>><br>
<br><br>>> As I said, I don't quite get why this does not cause at least image-side errors if not VM crashes in Squeak, but they should probably be fixed anyway :)<br>
<br><br>>><br>
<br><br>>> All the best,<br>
<br><br>>> Florin<br>
<br><br></blockquote></div>
<br><br><u></u><u></u><u></u></div></blockquote></div></blockquote></div>
<br><u></u><u></u><u></u><u></u><u></u></div></blockquote>
</div></div></blockquote></div>