diff for duplicates of <20210817125949.5bde9715@xps13> diff --git a/a/1.txt b/N1/1.txt index 7d19313..01943e3 100644 --- a/a/1.txt +++ b/N1/1.txt @@ -89,7 +89,8 @@ also have an answer or perhaps knows who to include in the discussion > > > jump to it. > > > > > -> > We still need some help and clarification from those who are very familiar with this sort of drivers or/and can test this particular driver. mxic_nfc_clk_enable() is the complementary function for mxic_nfc_clk_disable(). No other functions invoke clk_prepare_enable()/clk_disable_unprepare() in the driver. Unlikely somebody in its environment does that since driver specific clocks are dealt with. At the moment the driver invokes mxic_nfc_clk_disable() on error handling paths in probe, in remove and in mxic_nfc_set_freq(). mxic_nfc_clk_enable() is called just by mxic_nfc_set_freq() that moreover does this after calling mxic_nfc_clk_disable() first. So, we did not find any place in the driver that invokes mxic_nfc_clk_enable() prior to mxic_nfc_clk_disable(). Basing on this we added mxic_nfc_clk_enable() just after getting clocks. As I explained in the previous large e-mail, we may be wrong in our understanding of the driver environment or/and at specification of requirements being checked. It would be great if you will point out on our mistakes. +> > We still need some help and clarification from those who are very familiar with this sort of drivers or/and can test this particular driver. mxic_nfc_clk_enable() is the complementary function for mxic_nfc_clk_disable(). No other functions invoke clk_prepare_enable()/clk_disable_unprepare() in the driver. Unlikely somebody in its environment does that since driver specific clocks are dealt with. At the moment the driver invokes mxic_nfc_clk_disable() on error handling paths in probe, in remove and in mxic_nfc_set_freq(). mxic_nfc_clk_enable() is called just by mxic_nfc_set_freq() that moreover does this after calling mxic_nfc_clk_disable() first. So, we did not find any place in the driver that invokes mxic_nfc_clk_enable() prior to mxic_nfc_clk_disable(). Basing on this we added mxic_nfc_clk_enable() just after getting clocks. As I explained in the previous large e-mail, we may be wrong in our understanding of the driver environment or/and at specification of requirements being + checked. It would be great if you will point out on our mistakes. > > Enabling the clocks seems to only be needed to access the NAND device > and not the registers of the controller. Mason, is this statement @@ -107,7 +108,3 @@ also have an answer or perhaps knows who to include in the discussion > > Thanks, > Miquèl - -______________________________________________________ -Linux MTD discussion mailing list -http://lists.infradead.org/mailman/listinfo/linux-mtd/ diff --git a/a/content_digest b/N1/content_digest index 187044f..fc725d2 100644 --- a/a/content_digest +++ b/N1/content_digest @@ -108,7 +108,8 @@ "> > > jump to it.\n" "> > > \n" "> > \n" - "> > We still need some help and clarification from those who are very familiar with this sort of drivers or/and can test this particular driver. mxic_nfc_clk_enable() is the complementary function for mxic_nfc_clk_disable(). No other functions invoke clk_prepare_enable()/clk_disable_unprepare() in the driver. Unlikely somebody in its environment does that since driver specific clocks are dealt with. At the moment the driver invokes mxic_nfc_clk_disable() on error handling paths in probe, in remove and in mxic_nfc_set_freq(). mxic_nfc_clk_enable() is called just by mxic_nfc_set_freq() that moreover does this after calling mxic_nfc_clk_disable() first. So, we did not find any place in the driver that invokes mxic_nfc_clk_enable() prior to mxic_nfc_clk_disable(). Basing on this we added mxic_nfc_clk_enable() just after getting clocks. As I explained in the previous large e-mail, we may be wrong in our understanding of the driver environment or/and at specification of requirements being checked. It would be great if you will point out on our mistakes.\n" + "> > We still need some help and clarification from those who are very familiar with this sort of drivers or/and can test this particular driver. mxic_nfc_clk_enable() is the complementary function for mxic_nfc_clk_disable(). No other functions invoke clk_prepare_enable()/clk_disable_unprepare() in the driver. Unlikely somebody in its environment does that since driver specific clocks are dealt with. At the moment the driver invokes mxic_nfc_clk_disable() on error handling paths in probe, in remove and in mxic_nfc_set_freq(). mxic_nfc_clk_enable() is called just by mxic_nfc_set_freq() that moreover does this after calling mxic_nfc_clk_disable() first. So, we did not find any place in the driver that invokes mxic_nfc_clk_enable() prior to mxic_nfc_clk_disable(). Basing on this we added mxic_nfc_clk_enable() just after getting clocks. As I explained in the previous large e-mail, we may be wrong in our understanding of the driver environment or/and at specification of requirements being \n" + " checked. It would be great if you will point out on our mistakes.\n" "> \n" "> Enabling the clocks seems to only be needed to access the NAND device\n" "> and not the registers of the controller. Mason, is this statement\n" @@ -125,10 +126,6 @@ "> In all cases, the error path is wrong.\n" "> \n" "> Thanks,\n" - "> Miqu\303\250l\n" - "\n" - "______________________________________________________\n" - "Linux MTD discussion mailing list\n" - http://lists.infradead.org/mailman/listinfo/linux-mtd/ + "> Miqu\303\250l" -8370372a0f7e1856a3fa0434198f9282b9b4629aff209d8c5874691d2d0c28e5 +eba01975aba60b55c9e9ffab6dd2dc57a79f44a2845f7cd5fcdf9a8718b9e049
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.