sh: clk: Fix assignment from incompatible pointer type for ioreadX()
PATCH 1/1 sh: clk: Fix assignment from incompatible pointer type for ioreadX()
The ioreadX() helpers accept now pointer to const memory so declaration of read function needs updating.
This fixes build errors like:
drivers/sh/clk/cpg.c: In function ‘sh_clk_mstp_enable’:
drivers/sh/clk/cpg.c:49:9: error: assignment from incompatible pointer type [-Werror=incompatible-pointer-types]
read = ioread8;
---
Dear Andrew,
This was part of my v3 patchset commit 9ab7fb303cc1 ("iomap: Constify ioreadX() iomem argument (as in generic implementation)") but I think it was skipped when applying to your tree.
Maybe because it depends on commit 58c4d8659186 ("sh: clkfwk: remove r8/r16/r32") which landed later? Anyway it should go through your tree, I think.
--- drivers/sh/clk/cpg.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/sh/clk/cpg.c +1 -1
@@ -40,7 +40,7 @@ static int sh_clk_mstp_enable(struct clk *clk) {sh_clk_write (sh_clk_read (clk) & ~(1 << clk->enable_bit), clk);if (clk->status_reg) { -unsigned int (*read)(void __iomem *addr); +unsigned int (*read)(const void __iomem *addr);int i;void __iomem *mapped_status = (phys_addr_t )clk->status_reg - (phys_addr_t )clk->enable_reg + clk->mapped_reg;
Signature · 1 line
2.17.1
John Paul Adrian Glaubitz
maintainerHi Krzysztof!
Quoted 12 lines · Krzysztof Kozlowski
Maybe because it depends on commit
58c4d8659186("sh: clkfwk: remove r8/r16/r32") which landed later? Anyway it should go through your tree, I think.--- drivers/sh/clk/cpg.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)drivers/sh/clk/cpg.c +0
-0
drivers/sh/clk/cpg.c static int sh_clk_mstp_enable(struct clk *clk) · line 46
Quoted 8 lines
drivers/sh/clk/cpg.c +1
-1@@ -40,7 +40,7 @@ static int sh_clk_mstp_enable(struct clk *clk) {sh_clk_write (sh_clk_read (clk) & ~(1 << clk->enable_bit), clk);if (clk->status_reg) { -unsigned int (*read)(void __iomem *addr); +unsigned int (*read)(const void __iomem *addr);int i;void __iomem *mapped_status = (phys_addr_t )clk->status_reg - (phys_addr_t )clk->enable_reg + clk->mapped_reg;
Is that a reverted patch, by any chance? The patch does not apply for me and looking at the sources, the qualifier is already "const" [1].
Adrian
Quoted 1 line
[1] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/drivers/sh/clk/cpg.c#n58
Signature · 4 lines
.''`. John Paul Adrian Glaubitz : :' : Debian Developer - [email protected] `. `' Freie Universitaet Berlin - [email protected] `- GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913
Quoted 10 lines · John Paul Adrian Glaubitz
Hi Krzysztof!
Quoted 7 lines · Krzysztof Kozlowski
Maybe because it depends on commit
58c4d8659186("sh: clkfwk: remove r8/r16/r32") which landed later? Anyway it should go through your tree, I think.--- drivers/sh/clk/cpg.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
Quoted 16 lines
drivers/sh/clk/cpg.c static int sh_clk_mstp_enable(struct clk *clk) · line 46
Quoted 13 lines
drivers/sh/clk/cpg.c +1
-1@@ -40,7 +40,7 @@ static int sh_clk_mstp_enable(struct clk *clk) {sh_clk_write (sh_clk_read (clk) & ~(1 << clk->enable_bit), clk);if (clk->status_reg) { -unsigned int (*read)(void __iomem *addr); +unsigned int (*read)(const void __iomem *addr);int i;void __iomem *mapped_status = (phys_addr_t )clk->status_reg - (phys_addr_t )clk->enable_reg + clk->mapped_reg;Is that a reverted patch, by any chance? The patch does not apply for me and looking at the sources, the qualifier is already "const" [1].
The patch is for Andrew Morton's tree which contain two commits:
1. sh: clkfwk: remove r8/r16/r32
2. iomap: constify ioreadX() iomem argument (as in generic implementation)
This patch here was actually part of #2 because of being based on linux-next. However it disappeared when applying to Andrew's tree because patch #1 came later.
Best regards,
Krzysztof
John Paul Adrian Glaubitz
maintainerHi!
Quoted 7 lines · Krzysztof Kozlowski
The patch is for Andrew Morton's tree which contain two commits:
1. sh: clkfwk: remove r8/r16/r32
2. iomap: constifyioreadX()iomem argument (as in generic implementation)This patch here was actually part of #2 because of being based on linux-next. However it disappeared when applying to Andrew's tree because patch #1 came later.
Okay. I'll try to look at the other two patches later.
Adrian
Signature · 4 lines
.''`. John Paul Adrian Glaubitz : :' : Debian Developer - [email protected] `. `' Freie Universitaet Berlin - [email protected] `- GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913