memorder: Move \FloatBarrier to section end
PATCH 1/2 memorder: Move \FloatBarrier to section end
This \FloatBarrier is supposed to prevent floats of Section 15.2 from appearing in Section 15.3.
Move it to the end of Section 15.2
--- memorder/memorder.tex | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
memorder/memorder.tex +2 -2
@@ -2493,8 +2493,6 @@ In short, properly constructed release-acquire chains form a peaceful island of intuitive bliss surrounded by a strongly counter-intuitive sea of more complex memory-ordering constraints. -\FloatBarrier - \subsection{A Counter-Intuitive Case Study} \label{sec:memorder:A Counter-Intuitive Case Study}
@@ -2724,6 +2722,8 @@ Or perhaps more accurately, destroy some counter-productive intuitions. % @@@ Exercises? % @@@ Hardware details from Appendix? +\FloatBarrier + \section{Compile-Time Consternation} \label{sec:memorder:Compile-Time Consternation} %
base-commit: 7ac62b96acc23e612c1bb88389bc31b04215e402
Signature · 1 line
2.25.1
PATCH 2/2 memorder: Use desctiption list for PowerPC terminology
Using \paragraph under subsection level breaks normal document structure.
A description list should be a better fit for the purpose of presenting PowerPC terminology.
As a bonus, this change makes the end of the list of terminology becomes obvious to your eyes.
---
Hi Paul,
I think this is what you wanted do.
I changed the periods at the end of terms into colons. Do they look good to you?
Thanks, Akira
--
memorder/memorder.tex | 16 ++++++++++------ 1 file changed, 10 insertions(+), 6 deletions(-)
memorder/memorder.tex +10 -6
@@ -2571,15 +2571,17 @@ later \co{lwz} from \co{x} must precede \co{P0()}'s \co{stw} to \co{x}. Seeing how this can happen requires a rough understanding of the following PowerPC terminology. -\paragraph{Instruction commit.} +\begin{description}[style=nextline] + +\item[Instruction commit:] This can be thought of as the execution of that instruction as opposed to the memory-system consequences of having executed that instruction. -\paragraph{Write reaching coherence point.} +\item[Write reaching coherence point:] This can be thought of as the value written being deposited into the corresponding cache line. -\paragraph{Partial coherence commit.} +\item[Partial coherence commit:] This can be thought of as the system having worked out the order in which a pair of values written will be deposited into the corresponding cache line, but potentially well before that cache line arrives.
@@ -2589,7 +2591,7 @@ suggests that real PowerPC hardware does in fact use partial coherence commits to handle concurrent stores by multiple hardware threads within a single core. -\paragraph{Write propagate to thread.} +\item[Write propagate to thread:] This occurs when a second hardware thread becomes aware of the first hardware thread's write. The time at which a write propagates to a given thread might not have
@@ -2601,11 +2603,11 @@ the first thread's write's value might have been deposited into the corresponding cache line long before the second thread learns of that write. -\paragraph{Barrier propagate to thread.} +\item[Barrier propagate to thread:] Hardware threads make each other aware of memory-barrier instructions as needed by propagating them to each other. -\paragraph{Acknowledge \co{sync}.} +\item[Acknowledge \tco{sync}:] The PowerPC \co{sync} instruction implements the Linux kernel's \co{smp_mb()} full barrier. And one reason that the \co{sync} instruction provides such strong
@@ -2614,6 +2616,8 @@ threads, but these other threads must also acknowledge each \co{sync}. This two-way communication allows the hardware threads to cooperate to produce the required strong global ordering. +\end{description} + \begin{figure*}[tbp] \centering \resizebox{\textwidth}{!}{\includegraphics{memorder/PPCMEM0.png}}
Signature · 1 line
2.25.1
Re: [PATCH -perfbook 2/2] memorder: Use desctiption list for PowerPC terminology
Quoted 9 lines · Akira Yokosawa
Using \paragraph under subsection level breaks normal document structure.
A description list should be a better fit for the purpose of presenting PowerPC terminology.
As a bonus, this change makes the end of the list of terminology becomes obvious to your eyes.
Quoted 7 lines
Signed-off-by Akira Yokosawa <[email protected]>---
Hi Paul,I think this is what you wanted do.
I changed the periods at the end of terms into colons. Do they look good to you?
Looks good, queued and pushed, thank you!
I did not know about "style=nextline", very nice!
Good catch on \FloatBarrier! The reason I moved it was that while I was writing the new section, it was leaving most of a page blank. But now that that section is complete, it doesn't seem to be having much effect. So I removed it.
Please let me know if this removal had bad effects somewhere that I missed.
Thanx, Paul
Quoted 66 lines
Thanks, Akira
--memorder/memorder.tex | 16 ++++++++++------ 1 file changed, 10 insertions(+), 6 deletions(-)memorder/memorder.tex +10
-6@@ -2571,15 +2571,17 @@ later \co{lwz} from \co{x} must precede \co{P0()}'s \co{stw} to \co{x}. Seeing how this can happen requires a rough understanding of the following PowerPC terminology. -\paragraph{Instruction commit.} +\begin{description}[style=nextline] + +\item[Instruction commit:] This can be thought of as the execution of that instruction as opposed to the memory-system consequences of having executed that instruction. -\paragraph{Write reaching coherence point.} +\item[Write reaching coherence point:] This can be thought of as the value written being deposited into the corresponding cache line. -\paragraph{Partial coherence commit.} +\item[Partial coherence commit:] This can be thought of as the system having worked out the order in which a pair of values written will be deposited into the corresponding cache line, but potentially well before that cache line arrives.@@ -2589,7 +2591,7 @@ suggests that real PowerPC hardware does in fact use partial coherence commits to handle concurrent stores by multiple hardware threads within a single core. -\paragraph{Write propagate to thread.} +\item[Write propagate to thread:] This occurs when a second hardware thread becomes aware of the first hardware thread's write. The time at which a write propagates to a given thread might not have@@ -2601,11 +2603,11 @@ the first thread's write's value might have been deposited into the corresponding cache line long before the second thread learns of that write. -\paragraph{Barrier propagate to thread.} +\item[Barrier propagate to thread:] Hardware threads make each other aware of memory-barrier instructions as needed by propagating them to each other. -\paragraph{Acknowledge \co{sync}.} +\item[Acknowledge \tco{sync}:] The PowerPC \co{sync} instruction implements the Linux kernel's \co{smp_mb()} full barrier. And one reason that the \co{sync} instruction provides such strong@@ -2614,6 +2616,8 @@ threads, but these other threads must also acknowledge each \co{sync}. This two-way communication allows the hardware threads to cooperate to produce the required strong global ordering. +\end{description} + \begin{figure*}[tbp] \centering \resizebox{\textwidth}{!}{\includegraphics{memorder/PPCMEM0.png}}Signature · 1 line
2.25.1
Re: [PATCH -perfbook 2/2] memorder: Use desctiption list for PowerPC terminology
Quoted 14 lines · Paul E. McKenney
Quoted 13 lines · Akira Yokosawa
Using \paragraph under subsection level breaks normal document structure.
A description list should be a better fit for the purpose of presenting PowerPC terminology.
As a bonus, this change makes the end of the list of terminology becomes obvious to your eyes.
Signed-off-by Akira Yokosawa <[email protected]>---
Hi Paul,
Quoted 7 lines
Quoted 3 lines
I think this is what you wanted do.
I changed the periods at the end of terms into colons. Do they look good to you?Looks good, queued and pushed, thank you!
I did not know about "style=nextline", very nice!
It is provided by the "enumitem" package. I think there are several lists which is already using it.
There look to me other uses of \paragraph which can be converted to definition lists. I'll look in to them.
Quoted 8 lines
Good catch on \FloatBarrier! The reason I moved it was that while I was writing the new section, it was leaving most of a page blank. But now that that section is complete, it doesn't seem to be having much effect. So I removed it.
Please let me know if this removal had bad effects somewhere that I missed.
Commit abf7086e7959 ("memorder: Remove \FloatBarrier") removed the one in front of Section 15.6.2.
It has lost its relevance and, yes, you are right in removing it.
The one I was moving in patch 1/2 is the one in front of Section 15.2.8. It spoils the -1c builds and results in an almost blank page at the end of Section 15.2.7.
It turns out just removing the \FloatBarrier suffices.
Will send a patch v2 for the removal.
Thanks, Akira
Quoted 69 lines
Thanx, Paul
Quoted 66 lines
Thanks, Akira
--memorder/memorder.tex | 16 ++++++++++------ 1 file changed, 10 insertions(+), 6 deletions(-)memorder/memorder.tex +10
-6@@ -2571,15 +2571,17 @@ later \co{lwz} from \co{x} must precede \co{P0()}'s \co{stw} to \co{x}. Seeing how this can happen requires a rough understanding of the following PowerPC terminology. -\paragraph{Instruction commit.} +\begin{description}[style=nextline] + +\item[Instruction commit:] This can be thought of as the execution of that instruction as opposed to the memory-system consequences of having executed that instruction. -\paragraph{Write reaching coherence point.} +\item[Write reaching coherence point:] This can be thought of as the value written being deposited into the corresponding cache line. -\paragraph{Partial coherence commit.} +\item[Partial coherence commit:] This can be thought of as the system having worked out the order in which a pair of values written will be deposited into the corresponding cache line, but potentially well before that cache line arrives.@@ -2589,7 +2591,7 @@ suggests that real PowerPC hardware does in fact use partial coherence commits to handle concurrent stores by multiple hardware threads within a single core. -\paragraph{Write propagate to thread.} +\item[Write propagate to thread:] This occurs when a second hardware thread becomes aware of the first hardware thread's write. The time at which a write propagates to a given thread might not have@@ -2601,11 +2603,11 @@ the first thread's write's value might have been deposited into the corresponding cache line long before the second thread learns of that write. -\paragraph{Barrier propagate to thread.} +\item[Barrier propagate to thread:] Hardware threads make each other aware of memory-barrier instructions as needed by propagating them to each other. -\paragraph{Acknowledge \co{sync}.} +\item[Acknowledge \tco{sync}:] The PowerPC \co{sync} instruction implements the Linux kernel's \co{smp_mb()} full barrier. And one reason that the \co{sync} instruction provides such strong@@ -2614,6 +2616,8 @@ threads, but these other threads must also acknowledge each \co{sync}. This two-way communication allows the hardware threads to cooperate to produce the required strong global ordering. +\end{description} + \begin{figure*}[tbp] \centering \resizebox{\textwidth}{!}{\includegraphics{memorder/PPCMEM0.png}}Signature · 1 line
2.25.1
Re: [PATCH -perfbook 2/2] memorder: Use desctiption list for PowerPC terminology
Quoted 19 lines · Akira Yokosawa
Quoted 18 lines · Paul E. McKenney
Quoted 16 lines · Akira Yokosawa
Using \paragraph under subsection level breaks normal document structure.
A description list should be a better fit for the purpose of presenting PowerPC terminology.
As a bonus, this change makes the end of the list of terminology becomes obvious to your eyes.
Signed-off-by Akira Yokosawa <[email protected]>---
Hi Paul,I think this is what you wanted do.
I changed the periods at the end of terms into colons. Do they look good to you?
Quoted 6 lines
Quoted 3 lines
Looks good, queued and pushed, thank you!
I did not know about "style=nextline", very nice!
It is provided by the "enumitem" package. I think there are several lists which is already using it.
I am clearly not keeping up! ;-)
Quoted 2 lines
There look to me other uses of \paragraph which can be converted to definition lists. I'll look in to them.
That sounds very good, thank you!
Quoted 5 lines
Quoted 5 lines
Good catch on \FloatBarrier! The reason I moved it was that while I was writing the new section, it was leaving most of a page blank. But now that that section is complete, it doesn't seem to be having much effect. So I removed it.
Quoted 7 lines
Quoted 2 lines
Please let me know if this removal had bad effects somewhere that I missed.
Commit
abf7086e7959("memorder: Remove \FloatBarrier") removed the one in front of Section 15.6.2.It has lost its relevance and, yes, you are right in removing it.
Whew!!! ;-)
Quoted 7 lines
The one I was moving in patch 1/2 is the one in front of Section 15.2.8. It spoils the -1c builds and results in an almost blank page at the end of Section 15.2.7.
It turns out just removing the \FloatBarrier suffices.
Will send a patch v2 for the removal.
Very good, looking forward to it!
Thanx, Paul
Quoted 71 lines
Thanks, Akira
Quoted 69 lines
Thanx, Paul
Quoted 66 lines
Thanks, Akira
--memorder/memorder.tex | 16 ++++++++++------ 1 file changed, 10 insertions(+), 6 deletions(-)memorder/memorder.tex +10
-6@@ -2571,15 +2571,17 @@ later \co{lwz} from \co{x} must precede \co{P0()}'s \co{stw} to \co{x}. Seeing how this can happen requires a rough understanding of the following PowerPC terminology. -\paragraph{Instruction commit.} +\begin{description}[style=nextline] + +\item[Instruction commit:] This can be thought of as the execution of that instruction as opposed to the memory-system consequences of having executed that instruction. -\paragraph{Write reaching coherence point.} +\item[Write reaching coherence point:] This can be thought of as the value written being deposited into the corresponding cache line. -\paragraph{Partial coherence commit.} +\item[Partial coherence commit:] This can be thought of as the system having worked out the order in which a pair of values written will be deposited into the corresponding cache line, but potentially well before that cache line arrives.@@ -2589,7 +2591,7 @@ suggests that real PowerPC hardware does in fact use partial coherence commits to handle concurrent stores by multiple hardware threads within a single core. -\paragraph{Write propagate to thread.} +\item[Write propagate to thread:] This occurs when a second hardware thread becomes aware of the first hardware thread's write. The time at which a write propagates to a given thread might not have@@ -2601,11 +2603,11 @@ the first thread's write's value might have been deposited into the corresponding cache line long before the second thread learns of that write. -\paragraph{Barrier propagate to thread.} +\item[Barrier propagate to thread:] Hardware threads make each other aware of memory-barrier instructions as needed by propagating them to each other. -\paragraph{Acknowledge \co{sync}.} +\item[Acknowledge \tco{sync}:] The PowerPC \co{sync} instruction implements the Linux kernel's \co{smp_mb()} full barrier. And one reason that the \co{sync} instruction provides such strong@@ -2614,6 +2616,8 @@ threads, but these other threads must also acknowledge each \co{sync}. This two-way communication allows the hardware threads to cooperate to produce the required strong global ordering. +\end{description} + \begin{figure*}[tbp] \centering \resizebox{\textwidth}{!}{\includegraphics{memorder/PPCMEM0.png}}Signature · 1 line
2.25.1