| ↑back Search ←Prev date Next date→ Show only urls | (Click on time to select a line by its url) |
| Who | What | When |
|---|---|---|
| ezequielg | mripard: ah of course
mchehab: "[GIT PULL for v4.20-rc1] new experimental media request API" contains a patch from me, but it shouldn't be there. "media: dt-bindings: Document the Rockchip VPU bindings" should not be part of that pull. | [14:05] |
| mchehab | ezequielg: why not? it shouldn't go upstream? | [14:06] |
| ezequielg | no
the rockchip vpu is the JPEG encoder that we talked about you objected the JPEG_RAW format, and after some discussion I agreed with you. so I know have to re-do the driver the bindings are probably OK, but I wouldn't want to confuse people by submitting a bindings without a driver. | [14:07] |
| mchehab | ok, pls send me a revert patch. I'll add it to my tree and send later upstream | [14:08] |
| ezequielg | ok | [14:08] |
| mchehab | (I'll very likely send one last PR during the merge window - I'll place it there) | [14:09] |
| ezequielg | cool | [14:13] |
| ...... (idle for 29mn) | ||
| *** | prabhakarlad has left | [14:42] |
| ......................... (idle for 2h1mn) | ||
| jhautbois | pH5, I sent a mail on the ML as I can't get through an issue with mipi-csi2 with i.MX6... I don't think this is related to continuous clock, and am stuck. Thanks for the doc from yesterday, wad very helpful ! | [16:43] |
| pH5 | jhautbois: ok, I'll have a look later. | [16:44] |
| jhautbois | thanks | [16:47] |
| *** | benjiG has left | [16:48] |
| ...................................................... (idle for 4h26mn) | ||
| ezequielg | hverkuil: after our talk with pinchartl, do you think we can merge [PATCH v5 0/2] media: usb: pwc: Don't use coherent DMA buffers for ISO transfer ? | [21:14] |
| ...... (idle for 28mn) | ||
| pinchartl | ezequielg: I had a quick look at the patch. as explained before I think we need to fix a few things in the dma-mapping API in order to solve the problem cleanly, but given the performance improvement, I see no reason to block the patch
I'd love it if someone could work on the non-coherent allocation API though | [21:42] |
| ........................ (idle for 1h55mn) | ||
| ezequielg | pinchartl: well, I submitted a naive set for the DMA API
I have to say, I am also concerned about moving away from a CMA-backed path for the allocation and that's why I haven't done anything on stk1160, although a similar patch would improve performance on some popular platforms I just haven't made up my mind note that my objection is not about the patch itself, but about lack of clear directions. I know everyone hates parameters, but I think we could have just lived with some new parameter, and a comment saying it should go away, once X, Y, Z features get merged and merge the patch ages ago | [23:37] |
| ↑back Search ←Prev date Next date→ Show only urls | (Click on time to select a line by its url) |