pabs3 has quit [Remote host closed the connection]
pabs3 has joined #riscv
ldevulder_ is now known as ldevulder
mlw has quit [Ping timeout: 252 seconds]
mlw has joined #riscv
prabhakalad has quit [Remote host closed the connection]
prabhakalad has joined #riscv
<geist>
nice!
naoki has quit [Quit: naoki]
larzuk has joined #riscv
larzuk has quit [Ping timeout: 252 seconds]
larzuk has joined #riscv
larzuk has quit [Ping timeout: 255 seconds]
KREYREN_ has quit [Remote host closed the connection]
davidlt has joined #riscv
psydroid has joined #riscv
bitoff has quit [Ping timeout: 260 seconds]
larzuk has joined #riscv
jacklsw has quit [Ping timeout: 252 seconds]
<another|>
Whatever happened to horse creek?
<davidlt>
another|, no mass availability for it IIRC
JanC has quit [Ping timeout: 246 seconds]
<sorear>
very interesting that this is using a third party SoC. will be looking at volume and how they differentiate it from third party boards...
<davidlt>
well, this is the best
JanC has joined #riscv
<davidlt>
This is probably a true production SoC designed by ESWIN thus not some test chip. Sipeed said they will have SOM modules with P550 in May, probably the same chip.
Stat_headcrabed has joined #riscv
larzuk has quit [Ping timeout: 256 seconds]
<davidlt>
I doubt this is a cheap board, and overall Oasis probably is a better value (minus BMC).
<bjdooks>
iirc sophgo have a chip with similar cores too
* bjdooks
is sad to see horsecreek seems to be lame
<davidlt>
SG2380 / Oasis Gen 1
<davidlt>
They just announced another spec bump for the chip. It now has 256-bit memory bus, up to 128GB.
<davidlt>
They plan to use LPCAMM2 if it's available later this year. Still plan for Q4.
<unlord>
I have an SD card and df is showing 1.2G used on a partition where du shows 8.1M
<unlord>
how do I figure out what is happening?
markh has joined #riscv
<another|>
what fs?
<another|>
anything like transparent compression
<bjdooks>
unlord: could the fs have a stupidly big block size?
Stat_headcrabed has quit [Quit: Stat_headcrabed]
<unlord>
another|: ext4 I think, let me look
<unlord>
another|: yes, ext4
<another|>
hmm... sparse files?
davidlt has quit [Ping timeout: 240 seconds]
<sorear>
Very big disk with 1G allocated for the inode table/block bitmap/journal?
<bjdooks>
or there's blocks awaiting reclamation from a file that';s been deleted but not yet unreferecned?
<sorear>
or files that exist but aren't visible in the vfs or to du because you have something mounted on top of the directory?
<sorear>
gnu-compatible du matches df by default for sparse files, you have to use --apparent-size to get the other behavior
guerby_ is now known as guerby
davidlt has joined #riscv
JanC has quit [Read error: Connection reset by peer]
JanC has joined #riscv
JanC has quit [Excess Flood]
JanC has joined #riscv
tlwoerner_ has joined #riscv
tlwoerner has quit [Remote host closed the connection]
JanC has quit [Ping timeout: 256 seconds]
JanC has joined #riscv
unlord has quit [Changing host]
unlord has joined #riscv
luca_ has joined #riscv
luca_ is now known as OwlWizard
tlwoerner_ has quit [Quit: Leaving]
tlwoerner has joined #riscv
ntwk has quit [Quit: ntwk]
stolen has joined #riscv
Stat_headcrabed has joined #riscv
Stat_headcrabed has quit [Client Quit]
<arnd>
conchuod: I'm looking at the bugfix pull requests at the moment and I'm unused about "cache: sifive_ccache: Partially convert to a platform driver" in your drivers PR. Doesn't this mean that there cannnot be any DMA before the driver subsystem is up?
<arnd>
I wonder if that will cause even worse problems, e.g. if an early console driver relies on DMA
<conchuod>
arnd: Is the arch_initcall that registers the cache ops not sufficiently early?
Stat_headcrabed has joined #riscv
<arnd>
arch_initcall is about as early as drivers can be, but I think you can have DMA before that
<arnd>
I see that drivers/dma/*.c drivers are typically either arch_initcall or subsys_initcall, so at least anything using slave dma won't come before arch_initcall
<arnd>
conchuod: actually I guess the real problem is that the plic is now a device_initcall, which is much later than arch_initcall
<conchuod>
arnd: ye, that's the reason the patch was required to maintain the status quo w.r.t. dma.
<conchuod>
There's a bunch of drivers that do do core/postcore initcalls, if we do run into a device with really early DMA that needs this, there's at least options.
<arnd>
but then no other driver at {arch,subsys,fs,rootfs,device}_initcall level can rely on DMA to work
<conchuod>
If the cache ops are registered at arch level, why can't things after that use DMA?
<arnd>
the device_initcall() ones may or may not work, depending on the link order
<conchuod>
The part of the ccache driver that is now a platform driver is only the bits used for debugfs and by the edac driver.
<arnd>
ok, I guess that makes sense. So the irq is only required for EDAC, but managing the caches works without that?
<conchuod>
Yup
<arnd>
ok, I'll get this merged right away then, should send it out once my branch passes the build tests
<conchuod>
arnd: The interrupts are generated for corrected/uncorrected data errors in the l2.
<arnd>
I still have the feeling that an irqchip driver should be initialized earlier than device_initcall() time, because otherwise every device connected to the plic gets deferred probing, but that's not a problem with the cache driver patch
<arnd>
it looks like about half the platform drivers in drivers/irqchip are all builtin_platform_driver, the other ones open-code the registration using {core,postcore,arch,early}_initcall, so they are all over the place
JanC has quit [Ping timeout: 255 seconds]
BootLayer has joined #riscv
Noisytoot has quit [Remote host closed the connection]
Noisytoot has joined #riscv
OwlWizard has quit [Quit: OwlWizard]
handsome_feng has quit [Quit: Connection closed for inactivity]
Andre_Z has joined #riscv
davidlt has quit [Ping timeout: 240 seconds]
vagrantc has joined #riscv
BootLayer has quit [Quit: Leaving]
JanC has joined #riscv
knolle has quit [Quit: knolle]
knolle has joined #riscv
luca_ has joined #riscv
luca_ is now known as OwlWizard
KREYREN has joined #riscv
JanC has quit [Ping timeout: 246 seconds]
JanC has joined #riscv
JanC has quit [Ping timeout: 246 seconds]
KREYREN has quit [Remote host closed the connection]
KREYREN has joined #riscv
junaid_ has joined #riscv
stolen has quit [Quit: Connection closed for inactivity]
junaid_ has quit [Remote host closed the connection]
JanC has joined #riscv
JanC_ has joined #riscv
JanC is now known as Guest3202
JanC_ is now known as JanC
Stat_headcrabed has quit [Quit: Stat_headcrabed]
Guest3202 has quit [Ping timeout: 268 seconds]
davidlt has joined #riscv
davidlt has quit [Remote host closed the connection]
davidlt has joined #riscv
ln5_ has quit [Remote host closed the connection]
OwlWizard has quit [Quit: OwlWizard]
KREYREN has quit [Remote host closed the connection]
KREYREN has joined #riscv
Tenkawa has joined #riscv
jfsimon1981_c has quit [Remote host closed the connection]
freakazoid332 has joined #riscv
frkzoid has quit [Ping timeout: 268 seconds]
KREYREN has quit [Remote host closed the connection]
KREYREN_ has joined #riscv
KREYREN_ has quit [Remote host closed the connection]
KREYREN_ has joined #riscv
davidlt has quit [Ping timeout: 268 seconds]
KREYREN_ has quit [Remote host closed the connection]
KREYREN_ has joined #riscv
JanC has quit [Ping timeout: 268 seconds]
JanC has joined #riscv
KREYREN_ has quit [Remote host closed the connection]
hightower2 has joined #riscv
KREYREN has joined #riscv
KREYREN has quit [Remote host closed the connection]
KREYREN has joined #riscv
KREYREN has quit [Remote host closed the connection]
KREYREN has joined #riscv
KREYREN has quit [Remote host closed the connection]