Comments on: On FFA Chapter 2 http://ave1.org/2018/on-ffa-chapter-2/ Tue, 06 Oct 2020 17:03:05 +0000 http://polimedia.us hourly 1 By: A Walk among the Trees of V « Ossa Sepia http://ave1.org/2018/on-ffa-chapter-2/#comment-530 A Walk among the Trees of V « Ossa Sepia Fri, 13 Dec 2019 18:29:10 +0000 http://ave1.org/?p=53#comment-530 [...] very little towards no discussion published of *other people's code*. Ave1 has some on FFA (e.g. on chapter 2) and I have some on FFA as well (e.g. reading notes and an overview) but I can't say I saw much [...] [...] very little towards no discussion published of *other people's code*. Ave1 has some on FFA (e.g. on chapter 2) and I have some on FFA as well (e.g. reading notes and an overview) but I can't say I saw much [...]

]]> By: ave1 http://ave1.org/2018/on-ffa-chapter-2/#comment-31 ave1 Wed, 14 Feb 2018 10:47:39 +0000 http://ave1.org/?p=53#comment-31 Well I did say "I see no use", which is a statement about my limited ability to foresee this field, not about the procedures. The only problem with "I see no use", is that it makes it harder for <em>me</em> to evaluate their correctness. I can see these correctly do their actions on the first element of the FZ, I cannot (i.e. my limit) see why not also for tail or anywhere else. One of the comments I see in channel about pgp / mpi, is that is has these functions that seem dead but are used and may contain a fatal error, which can be related to the difference between "forseen use" (20 or more years ago) and "actual use" (a couple of years ago when somebody just looked at the name of that function an thought it would fit or something). Well I did say "I see no use", which is a statement about my limited ability to foresee this field, not about the procedures.

The only problem with "I see no use", is that it makes it harder for me to evaluate their correctness.

I can see these correctly do their actions on the first element of the FZ, I cannot (i.e. my limit) see why not also for tail or anywhere else.

One of the comments I see in channel about pgp / mpi, is that is has these functions that seem dead but are used and may contain a fatal error, which can be related to the difference between "forseen use" (20 or more years ago) and "actual use" (a couple of years ago when somebody just looked at the name of that function an thought it would fit or something).

]]> By: Stanislav Datskovskiy http://ave1.org/2018/on-ffa-chapter-2/#comment-29 Stanislav Datskovskiy Mon, 12 Feb 2018 17:07:35 +0000 http://ave1.org/?p=53#comment-29 I must add, most of the "useless" knobs do in fact come into use in later chapters. (It is possible that some are left without any conceivable use; these will be removed.) Their presence is an artifact of the ongoing rewrite (the bulk of the material in fact existed, in a rougher draft form, since September.) I must add, most of the "useless" knobs do in fact come into use in later chapters. (It is possible that some are left without any conceivable use; these will be removed.)

Their presence is an artifact of the ongoing rewrite (the bulk of the material in fact existed, in a rougher draft form, since September.)

]]> By: ave1 http://ave1.org/2018/on-ffa-chapter-2/#comment-28 ave1 Mon, 12 Feb 2018 13:20:54 +0000 http://ave1.org/?p=53#comment-28 Yes, indeed it refers to the option described before with the 2d array. (I'll make a small update to the paragraph to make the sentence clearer!) The version implemented in FFA works perfectly and has exact the same code path and memory access pattern regardless of Sel. Yes, indeed it refers to the option described before with the 2d array. (I'll make a small update to the paragraph to make the sentence clearer!)

The version implemented in FFA works perfectly and has exact the same code path and memory access pattern regardless of Sel.

]]> By: Stanislav Datskovskiy http://ave1.org/2018/on-ffa-chapter-2/#comment-27 Stanislav Datskovskiy Mon, 12 Feb 2018 13:11:27 +0000 http://ave1.org/?p=53#comment-27 or hmm, looks like text referred to a hypothetical alternative? or hmm, looks like text referred to a hypothetical alternative?

]]> By: Stanislav Datskovskiy http://ave1.org/2018/on-ffa-chapter-2/#comment-26 Stanislav Datskovskiy Mon, 12 Feb 2018 13:08:09 +0000 http://ave1.org/?p=53#comment-26 > Unfortunately, this code accesses memory ever so slightly differently based on the Sel bit, and thus might leak information. This is nonsense, there is no change in the addressing. From where did you get the notion? > Unfortunately, this code accesses memory ever so slightly differently based on the Sel bit, and thus might leak information.

This is nonsense, there is no change in the addressing.
From where did you get the notion?

]]>