kjn260 has quit [Remote host closed the connection]
warpme has joined #linux-sunxi
apritzel has quit [Ping timeout: 480 seconds]
jkm has quit [Quit: leaving]
utsweetyfish has quit [autokilled: Host violated network policy. - Contact support@oftc.net for help. (2024-06-14 06:56:32)]
kjn260 has joined #linux-sunxi
Schimsalabim has quit [Ping timeout: 480 seconds]
Schimsalabim has joined #linux-sunxi
mripard_ has joined #linux-sunxi
mripard is now known as Guest9588
mripard_ is now known as mripard
Guest9588 has quit [Ping timeout: 480 seconds]
Schimsalabim has quit [Read error: Connection reset by peer]
Schimsalabim has joined #linux-sunxi
utsweetyfish has joined #linux-sunxi
Schimsalabim has quit [Ping timeout: 480 seconds]
Schimsalabim has joined #linux-sunxi
Schimsalabim has quit [Read error: Connection reset by peer]
Schimsalabim has joined #linux-sunxi
warpme has quit []
Schimsalabim has quit [Read error: No route to host]
Schimsalabim has joined #linux-sunxi
apritzel has joined #linux-sunxi
warpme has joined #linux-sunxi
wa has quit [Remote host closed the connection]
<loki666>
apritzel I'll test your hotfix later
<apritzel>
loki666: thanks! Please use pure mainline, plus this patch. I sincerely hope it works, as Lee needed some pushing to actually take it for v6.10 still
vpeter has quit [Read error: Connection reset by peer]
mripard has quit [Quit: mripard]
wingrime1 has quit []
wingrime-ww has joined #linux-sunxi
wingrime-ww has quit []
Robot_ has joined #linux-sunxi
electricworry has joined #linux-sunxi
Schimsalabim has quit [Ping timeout: 480 seconds]
Schimsalabim has joined #linux-sunxi
vpeter has joined #linux-sunxi
JohnDoe_71Rus has joined #linux-sunxi
thecheofusi has joined #linux-sunxi
thecheofusi has quit []
thecheofusi has joined #linux-sunxi
<thecheofusi>
Suppose I have a non contiguous buffer in memory whose size is 1,536,000 bytes, and the sunxi iommu
<thecheofusi>
I call dma_map_sgtable on this buffer, and the iommu maps the whole thing into 1 contiguous dma address space
<thecheofusi>
Then I unmap with dma_unmap_sgtable(), and after some time call dma_map_sgtable() again
<thecheofusi>
Is there any reason why the second call to dma_map_sgtable() always returns a region in dma address space that is non contiguous?
<thecheofusi>
I can guarantee that no other device is using the iommu
<jakllsch>
depends
<thecheofusi>
jakllsch: Could you be more specific?
<jakllsch>
could the iommu space have gotten fragmented in the interim?
<loki666>
apritzel: patch is working, you want me to reply on ML? looks like Lee didn't wait for it
<apritzel>
loki666: many thanks for testing! Indeed seem like he just went ahead (good!), but you can reply for the records anyway, if you like
<loki666>
ok
<thecheofusi>
jakllsch: don't all mappings get cleaned up after dma_unmap_sgtable ?
<loki666>
apritzel: what's the use of the rest of the series, it seems to be related to running on battery, but I tested on battery, and it works
<apritzel>
it's an attempt to model the built-in boost regulator, which brings the battery voltage up to 5.0V for the USB bus
<apritzel>
not sure how your DT looks like, but the mainline DT doesn't reference that (fortunately, otherwise it would be broken again)
<apritzel>
macromorgan was messing around with that, but something was funny there, IIRC
linkmauve has left #linux-sunxi [Mise à jour !]
<thecheofusi>
jakllsch: I figured out what the problem was
<thecheofusi>
turns out I forgot to set the DMA max segment size
<thecheofusi>
So the DMA-mapping code was splitting the mapping into chunks of a random (?) size that hovered around 60K
<apritzel>
thecheofusi: what SoC is this on? Curious if your board has 4GB of DRAM...
<thecheofusi>
t113-s3
<apritzel>
ah, surely not 4GB then ;-)
linkmauve has joined #linux-sunxi
<thecheofusi>
why do you ask?
thecheofusi has quit [Remote host closed the connection]
<apritzel>
because the IOMMU in the H616 (and later) is still limited to 4GB of address space, both for input and output addresses, so cannot cover the last GB of DRAM on devices with 4GB
electricworry has quit [Remote host closed the connection]
apritzel has quit [Ping timeout: 480 seconds]
Nosnibor has joined #linux-sunxi
Schimsalabim has quit [Read error: Connection reset by peer]
Schimsalabim has joined #linux-sunxi
Schimsalabim has quit [Read error: Connection reset by peer]
Schimsalabim has joined #linux-sunxi
Nosnibor has quit []
JohnDoe_71Rus has quit [Quit: KVIrc KVIrc Quasar 5.2.2, revision: 5.2.0+git-7584-200e7bcec, build type: debug, sources date: 20160102, built on: 2024-05-13 20:04:01 UTC 5.2.0+git-7584-]
apritzel has joined #linux-sunxi
warpme has quit []
hentai has quit [Remote host closed the connection]