| Message ID | 20210826025144.51992-1-chunfeng.yun@mediatek.com |
|---|---|
| State | New |
| Headers | show |
| Series | [next,v2,1/6] Revert "usb: xhci-mtk: relax TT periodic bandwidth allocation" | expand |
Hi Greg, On Thu, 2021-08-26 at 13:41 +0200, Greg Kroah-Hartman wrote: > On Thu, Aug 26, 2021 at 10:51:39AM +0800, Chunfeng Yun wrote: > > As discussed in following patch: > > https://patchwork.kernel.org/patch/12420339 > > > > No need calculate number of uframes again, but should use value > > form check_sch_tt(), if we plan to remove extra CS, also can do > > it in check_sch_tt(). So revert this patch, and prepare to send > > new patch for it. > > > > This reverts commit 548011957d1d72e0b662300c8b32b81d593b796e. > > > > Cc: Ikjoon Jang <ikjn@chromium.org> > > Signed-off-by: Chunfeng Yun <chunfeng.yun@mediatek.com> > > --- > > v2: no changes > > This series does not apply to my tree at all now, can you please > rebase > and resend? Very sorry, I send out two series [1][2] for xhci-mtk, but don't take care of conflicts, suppose that series [1] will be applied firstly, due to one binding patch [3] of series [2] is not acked/reviewed by Rob (I think only need modify some misleading commit message). anyway, I find that only one patch [4] of series [1] is not applied, so I'll fix conficts and resend it based on usb-testing branch. Sorry again. [1]: Series = [next,v2,1/6] Revert "usb: xhci-mtk: relax TT periodic bandwidth allocation" https://patchwork.kernel.org/project/linux-mediatek/list/?series=537471 [2]: Series = [RESEND,1/9] dt-bindings: usb: mtk-xhci: add optional property to disable usb2 ports https://patchwork.kernel.org/project/linux-mediatek/list/?series=532595 [3]: https://patchwork.kernel.org/project/linux-mediatek/patch/1629189389-18779-2-git-send-email-chunfeng.yun@mediatek.com/ [RESEND,2/9] dt-bindings: usb: mtk-xhci: add compatible for mt8195 [4]: https://patchwork.kernel.org/project/linux-mediatek/patch/20210826025144.51992-6-chunfeng.yun@mediatek.com/ [next,v2,6/6] usb: xhci-mtk: allow bandwidth table rollover > > thanks, > > greg k-h
Hi Greg, On Thu, 2021-08-26 at 13:41 +0200, Greg Kroah-Hartman wrote: > On Thu, Aug 26, 2021 at 10:51:39AM +0800, Chunfeng Yun wrote: > > As discussed in following patch: > > https://patchwork.kernel.org/patch/12420339 > > > > No need calculate number of uframes again, but should use value > > form check_sch_tt(), if we plan to remove extra CS, also can do > > it in check_sch_tt(). So revert this patch, and prepare to send > > new patch for it. > > > > This reverts commit 548011957d1d72e0b662300c8b32b81d593b796e. > > > > Cc: Ikjoon Jang <ikjn@chromium.org> > > Signed-off-by: Chunfeng Yun <chunfeng.yun@mediatek.com> > > --- > > v2: no changes > > This series does not apply to my tree at all now, can you please > rebase > and resend? Very sorry, I send two series [1][2] for xhci-mtk, but don't take care of conficts: I suppose you will apply [1] firstly, due to [3] is not acked or reviewed by Rob (I think, only need modify commit message). Anyway, only one patch [4] is not in usb-testing branch due to conflicts, I'll fix it and send out, sorry again. [1]. Series = [next,v2,1/6] Revert "usb: xhci-mtk: relax TT periodic bandwidth allocation" https://patchwork.kernel.org/project/linux-mediatek/list/?series=537471 [2]. Series = [RESEND,1/9] dt-bindings: usb: mtk-xhci: add optional property to disable usb2 ports https://patchwork.kernel.org/project/linux-mediatek/list/?series=532595 [3]: https://patchwork.kernel.org/project/linux-mediatek/patch/1629189389-18779-2-git-send-email-chunfeng.yun@mediatek.com/ [RESEND,2/9] dt-bindings: usb: mtk-xhci: add compatible for mt8195 [4]: https://patchwork.kernel.org/project/linux-mediatek/patch/20210826025144.51992-6-chunfeng.yun@mediatek.com/ [next,v2,6/6] usb: xhci-mtk: allow bandwidth table rollover > > thanks, > > greg k-h
diff --git a/drivers/usb/host/xhci-mtk-sch.c b/drivers/usb/host/xhci-mtk-sch.c index 46cbf5d54f4f..f9b4d27ce449 100644 --- a/drivers/usb/host/xhci-mtk-sch.c +++ b/drivers/usb/host/xhci-mtk-sch.c @@ -459,17 +459,16 @@ static int check_fs_bus_bw(struct mu3h_sch_ep_info *sch_ep, int offset) u32 num_esit, tmp; int base; int i, j; - u8 uframes = DIV_ROUND_UP(sch_ep->maxpkt, FS_PAYLOAD_MAX); num_esit = XHCI_MTK_MAX_ESIT / sch_ep->esit; - - if (sch_ep->ep_type == INT_IN_EP || sch_ep->ep_type == ISOC_IN_EP) - offset++; - for (i = 0; i < num_esit; i++) { base = offset + i * sch_ep->esit; - for (j = 0; j < uframes; j++) { + /* + * Compared with hs bus, no matter what ep type, + * the hub will always delay one uframe to send data + */ + for (j = 0; j < sch_ep->cs_count; j++) { tmp = tt->fs_bus_bw[base + j] + sch_ep->bw_cost_per_microframe; if (tmp > FS_PAYLOAD_MAX) return -ESCH_BW_OVERFLOW; @@ -547,8 +546,6 @@ static void update_sch_tt(struct mu3h_sch_ep_info *sch_ep, bool used) u32 base, num_esit; int bw_updated; int i, j; - int offset = sch_ep->offset; - u8 uframes = DIV_ROUND_UP(sch_ep->maxpkt, FS_PAYLOAD_MAX); num_esit = XHCI_MTK_MAX_ESIT / sch_ep->esit; @@ -557,13 +554,10 @@ static void update_sch_tt(struct mu3h_sch_ep_info *sch_ep, bool used) else bw_updated = -sch_ep->bw_cost_per_microframe; - if (sch_ep->ep_type == INT_IN_EP || sch_ep->ep_type == ISOC_IN_EP) - offset++; - for (i = 0; i < num_esit; i++) { - base = offset + i * sch_ep->esit; + base = sch_ep->offset + i * sch_ep->esit; - for (j = 0; j < uframes; j++) + for (j = 0; j < sch_ep->cs_count; j++) tt->fs_bus_bw[base + j] += bw_updated; }
As discussed in following patch: https://patchwork.kernel.org/patch/12420339 No need calculate number of uframes again, but should use value form check_sch_tt(), if we plan to remove extra CS, also can do it in check_sch_tt(). So revert this patch, and prepare to send new patch for it. This reverts commit 548011957d1d72e0b662300c8b32b81d593b796e. Cc: Ikjoon Jang <ikjn@chromium.org> Signed-off-by: Chunfeng Yun <chunfeng.yun@mediatek.com> --- v2: no changes --- drivers/usb/host/xhci-mtk-sch.c | 20 +++++++------------- 1 file changed, 7 insertions(+), 13 deletions(-)