mumptai has quit [Quit: Verlassend]
sgstair_ has joined ##openfpga
sgstair has quit [Ping timeout: 260 seconds]
emeb has quit [Quit: Leaving.]
mwk has quit [Ping timeout: 272 seconds]
Bike has quit [Quit: Lost terminal]
mwk has joined ##openfpga
<promach3> hell__: why need to postpone ?
X-Scale` has joined ##openfpga
NorthStar` has joined ##openfpga
X-Scale has quit [Ping timeout: 272 seconds]
X-Scale` is now known as X-Scale
NorthStar` has quit [Quit: HydraIRC -> http://www.hydrairc.com <- s0 d4Mn l33t |t'z 5c4rY!]
sgstair_ is now known as sgstair
Degi_ has joined ##openfpga
Degi has quit [Ping timeout: 264 seconds]
Degi_ is now known as Degi
genii has quit [Quit: sleepytimes now]
s_frit has quit [Remote host closed the connection]
s_frit has joined ##openfpga
srk has quit [Remote host closed the connection]
srk has joined ##openfpga
craigjb has joined ##openfpga
craigjb has quit [Quit: ZNC 1.7.3 - https://znc.in]
craigjb has joined ##openfpga
emeb_mac has quit [Quit: Leaving.]
jeanthom has joined ##openfpga
sorki has joined ##openfpga
srk has quit [Ping timeout: 240 seconds]
sorki is now known as srk
mumptai has joined ##openfpga
jeanthom has quit [Ping timeout: 256 seconds]
jeanthom has joined ##openfpga
<hell__> promach3: it can improve performance
<promach3> hell__: huh ?
<hell__> you can't service read/write requests while the memory is refreshing
<hell__> so your memory controller's scheduler could decide to postpone a few refresh cycles to complete a request earlier
<hell__> and do the postponed refresh cycles afterwards, when it is idle
<promach3> but what is the relationship between 8 refresh commands and 9x tREFI ?
<hell__> fence posts = refresh commands
<hell__> fence sections = tREFI intervals
<promach3> yes, but we are not talking about fence post
<promach3> wait
<hell__> grab a piece of paper, draw a fence with 10 posts and 9 sections
<hell__> JEDEC says the maximum time between two refresh commands is 9x tREFI. Our fence is 9x tREFI long, so we can keep only the first and last posts, and we would still be fine
<promach3> hell__: refresh commands != fence post
<hell__> that is, we can postpone 10 - 2 = 8 refresh commands
<promach3> each refresh command spans across 1 tREFI interval
<promach3> please correct me if wrong
<promach3> why 10 -2 ?
<hell__> no, refresh commands span one tRFC
<hell__> for simplicity, I'm assuming tRFC is negligible in this example
<hell__> the 10 - 2 comes from the previous reasoning
<promach3> in this case, fence section = tRFC interval
<promach3> Why fence sections = tREFI intervals ?
<hell__> because we were talking about tREFI, weren't we?
* hell__ sighs
<promach3> wait, let me read the diagram again
s_frit_ has joined ##openfpga
<promach3> hell__: I understand it now
s_frit has quit [Ping timeout: 272 seconds]
<hell__> I really need a virtual blackboard :D
<promach3> What about self-refresh command ?
mumptai has quit [Remote host closed the connection]
Bike has joined ##openfpga
emeb has joined ##openfpga
genii has joined ##openfpga
s_frit has joined ##openfpga
s_frit_ has quit [Ping timeout: 264 seconds]
emeb has quit [Ping timeout: 246 seconds]
emeb has joined ##openfpga
Raito_Bezarius has quit [Ping timeout: 260 seconds]
peeps[zen] has joined ##openfpga
peepsalot has quit [Ping timeout: 272 seconds]
peepsalot has joined ##openfpga
peeps[zen] has quit [Ping timeout: 264 seconds]
tmeissner has joined ##openfpga
emeb_mac has joined ##openfpga
Raito_Bezarius has joined ##openfpga
SpaceCoaster_ has quit [Quit: ZNC 1.7.2+deb3 - https://znc.in]
SpaceCoaster has joined ##openfpga
mumptai has joined ##openfpga
Maylay has quit [Ping timeout: 240 seconds]
Maylay has joined ##openfpga
Maylay has quit [Ping timeout: 246 seconds]
tmeissner has quit [Quit: My MacBook Air has gone to sleep. ZZZzzz…]
tmeissner has joined ##openfpga
tmeissner has quit [Client Quit]
Maylay has joined ##openfpga
mumptai has quit [Quit: Verlassend]
s_frit has quit [Ping timeout: 240 seconds]
emeb has quit [Quit: Leaving.]
jeanthom has quit [Ping timeout: 240 seconds]
jn__ has quit [Ping timeout: 272 seconds]
jn__ has joined ##openfpga