RE: [PATCH] Revert "mmc: sdhci: set to INT_DATA_END when there are data"

2021-07-06 Thread Andy.Wu
Hi All Since patch also reviewed by Jaehoon, how about merge it into mainline? Best Regards Andy Wu > -Original Message- > From: Jaehoon Chung > Sent: Thursday, May 20, 2021 6:03 AM > To: Wu, Andy ; peng@nxp.com; > jh80.ch...@gmail.com; u-boot@lists.denx.de > Cc: CPGS > Subject: Re

RE: [PATCH] Revert "mmc: sdhci: set to INT_DATA_END when there are data"

2021-05-11 Thread Andy.Wu
Hi Jaehoon > If you're ok, I will test after reverted the patch on tomorrow, and I will > share > result. > Or I will try to reproduce timeout issue on 410c board. Sorry, but is there any update for this comments? Best Regards Andy Wu > -Original Message- > From: U-Boot On Behalf Of J

RE: [PATCH] Revert "mmc: sdhci: set to INT_DATA_END when there are data"

2021-03-22 Thread Andy.Wu
Hi Jaehoon > Did you test on latest u-boot? v2018.01 was too old version. > Yes, we tested on v2020.04, although there is no such issue, but I think it just depends on call sequence timing. > And if my understanding is right, INT_DATA_END needs to set when there is a > data. If there is no da

RE: [PATCH] Revert "mmc: sdhci: set to INT_DATA_END when there are data"

2021-03-17 Thread Andy.Wu
Hi > I don't want to revert this commit. Is there any issue without this? Without revert commit 17ea3c86, Some board, like Dragonboard 410c will meet transfer data timeout error (we used v2018.01): U-Boot 2018.01 (Nov 26 2020 - 03:31:09 +) Qualcomm-DragonBoard 410C DRAM: 986 MiB MMC: sd

RE: [PATCH] Revert "mmc: sdhci: set to INT_DATA_END when there are data"

2021-03-17 Thread Andy.Wu
Reviewed-by: Andy Wu Best Regards Andy Wu > -Original Message- > From: U-Boot On Behalf Of > yuezhang...@sony.com > Sent: Wednesday, March 17, 2021 2:45 PM > To: u-boot@lists.denx.de > Cc: peng@nxp.com; pa...@antoniou-consulting.com > Subject: [PATCH] Revert "mmc: sdhci: set to INT_

RE: [PATCH] autoboot: fix illegal memory access when stop key and delay key are empty

2021-01-17 Thread Andy.Wu
> > You could avoid the subtraction instead of changing the type: > > > > -for (i = 0; i < presskey_max - 1; i++) > > +for (i = 0; i + 1 < presskey_max; i++) > This style seems not typically way for for loop, how do you think? I found one similar for loop style in u-boot source code, it seems aim t

RE: [PATCH] autoboot: fix illegal memory access when stop key and delay key are empty

2021-01-17 Thread Andy.Wu
> Both indices cannot be negative. So it is understandable that u_int was > chosen. Yes, u_int is understandable that the length is never be negative, but define the length parameter as "int" also usable. Some example in current u-boot source code: $ grep -rn length common/ | grep "int " common/

Re: [PATCH] mmc: sdhci: skip cache invalidation if DMA is not used

2021-01-14 Thread Andy.Wu
The config usage keep the same style as function sdhci_prepare_dma(). Reviewed-by: Andy Wu > +#if (defined(CONFIG_MMC_SDHCI_SDMA) || CONFIG_IS_ENABLED(MMC_SDHCI_ADMA)) > dma_unmap_single(host->start_addr, data->blocks * data->blocksize, > mmc_get_dma_dir(data)); > +#endif -- Sent from: