From 021376a3b655364c92c10be544a3319946a792e8 Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Wed, 13 May 2009 20:30:24 +0000 Subject: powerpc/ftrace: Use pr_devel() in ftrace.c pr_debug() can now result in code being generated even when #DEBUG is not defined. That's not really desirable in the ftrace code which we want to be snappy. With CONFIG_DYNAMIC_DEBUG=y: size before: text data bss dec hex filename 3334 672 4 4010 faa arch/powerpc/kernel/ftrace.o size after: text data bss dec hex filename 2616 360 4 2980 ba4 arch/powerpc/kernel/ftrace.o Signed-off-by: Michael Ellerman Acked-by: Steven Rostedt Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/ftrace.c | 20 ++++++++++---------- 1 file changed, 10 insertions(+), 10 deletions(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/ftrace.c b/arch/powerpc/kernel/ftrace.c index 70e2a736be1..5b078ee391f 100644 --- a/arch/powerpc/kernel/ftrace.c +++ b/arch/powerpc/kernel/ftrace.c @@ -157,7 +157,7 @@ __ftrace_make_nop(struct module *mod, * 0xe8, 0x4c, 0x00, 0x28, ld r2,40(r12) */ - pr_debug("ip:%lx jumps to %lx r2: %lx", ip, tramp, mod->arch.toc); + pr_devel("ip:%lx jumps to %lx r2: %lx", ip, tramp, mod->arch.toc); /* Find where the trampoline jumps to */ if (probe_kernel_read(jmp, (void *)tramp, sizeof(jmp))) { @@ -165,7 +165,7 @@ __ftrace_make_nop(struct module *mod, return -EFAULT; } - pr_debug(" %08x %08x", jmp[0], jmp[1]); + pr_devel(" %08x %08x", jmp[0], jmp[1]); /* verify that this is what we expect it to be */ if (((jmp[0] & 0xffff0000) != 0x3d820000) || @@ -181,18 +181,18 @@ __ftrace_make_nop(struct module *mod, offset = ((unsigned)((unsigned short)jmp[0]) << 16) + (int)((short)jmp[1]); - pr_debug(" %x ", offset); + pr_devel(" %x ", offset); /* get the address this jumps too */ tramp = mod->arch.toc + offset + 32; - pr_debug("toc: %lx", tramp); + pr_devel("toc: %lx", tramp); if (probe_kernel_read(jmp, (void *)tramp, 8)) { printk(KERN_ERR "Failed to read %lx\n", tramp); return -EFAULT; } - pr_debug(" %08x %08x\n", jmp[0], jmp[1]); + pr_devel(" %08x %08x\n", jmp[0], jmp[1]); ptr = ((unsigned long)jmp[0] << 32) + jmp[1]; @@ -269,7 +269,7 @@ __ftrace_make_nop(struct module *mod, * 0x4e, 0x80, 0x04, 0x20 bctr */ - pr_debug("ip:%lx jumps to %lx", ip, tramp); + pr_devel("ip:%lx jumps to %lx", ip, tramp); /* Find where the trampoline jumps to */ if (probe_kernel_read(jmp, (void *)tramp, sizeof(jmp))) { @@ -277,7 +277,7 @@ __ftrace_make_nop(struct module *mod, return -EFAULT; } - pr_debug(" %08x %08x ", jmp[0], jmp[1]); + pr_devel(" %08x %08x ", jmp[0], jmp[1]); /* verify that this is what we expect it to be */ if (((jmp[0] & 0xffff0000) != 0x3d600000) || @@ -293,7 +293,7 @@ __ftrace_make_nop(struct module *mod, if (tramp & 0x8000) tramp -= 0x10000; - pr_debug(" %lx ", tramp); + pr_devel(" %lx ", tramp); if (tramp != addr) { printk(KERN_ERR @@ -402,7 +402,7 @@ __ftrace_make_call(struct dyn_ftrace *rec, unsigned long addr) /* ld r2,40(r1) */ op[1] = 0xe8410028; - pr_debug("write to %lx\n", rec->ip); + pr_devel("write to %lx\n", rec->ip); if (probe_kernel_write((void *)ip, op, MCOUNT_INSN_SIZE * 2)) return -EPERM; @@ -442,7 +442,7 @@ __ftrace_make_call(struct dyn_ftrace *rec, unsigned long addr) return -EINVAL; } - pr_debug("write to %lx\n", rec->ip); + pr_devel("write to %lx\n", rec->ip); if (probe_kernel_write((void *)ip, &op, MCOUNT_INSN_SIZE)) return -EPERM; -- cgit v1.2.3 From c3cf8667ed7db58c1960958cbb0a9098d513cc60 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Fri, 15 May 2009 04:33:54 +0000 Subject: powerpc/ftrace: Fix constraint to be early clobber After upgrading my distcc boxes from gcc 4.2.2 to 4.4.0, the function graph tracer broke. This was discovered on my x86 boxes. The issue is that gcc used the same register for an output as it did for an input in an asm statement. I first thought this was a bug in gcc and reported it. I was notified that gcc was correct and that the output had to be flagged as an "early clobber". I noticed that powerpc had the same issue and this patch fixes it. Signed-off-by: Steven Rostedt Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/ftrace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/ftrace.c b/arch/powerpc/kernel/ftrace.c index 5b078ee391f..2d182f119d1 100644 --- a/arch/powerpc/kernel/ftrace.c +++ b/arch/powerpc/kernel/ftrace.c @@ -594,7 +594,7 @@ void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr) PPC_LONG "2b,4b\n" ".previous" - : [old] "=r" (old), [faulted] "=r" (faulted) + : [old] "=&r" (old), [faulted] "=r" (faulted) : [parent] "r" (parent), [return_hooker] "r" (return_hooker) : "memory" ); -- cgit v1.2.3 From 0e337b42d620ca7c45fe64e64dd71957c56216c9 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Sun, 17 May 2009 18:29:03 +0000 Subject: powerpc: Explicit alignment for .data.cacheline_aligned I don't think anything guarantees that the objects in data.page_aligned are a multiple of PAGE_SIZE, thus the section may end on any boundary. So the following section, .data.cacheline_aligned needs an explicit alignment. Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/vmlinux.lds.S | 1 + 1 file changed, 1 insertion(+) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/vmlinux.lds.S b/arch/powerpc/kernel/vmlinux.lds.S index a047a6cfca4..8ef8a14abc9 100644 --- a/arch/powerpc/kernel/vmlinux.lds.S +++ b/arch/powerpc/kernel/vmlinux.lds.S @@ -264,6 +264,7 @@ SECTIONS *(.data.page_aligned) } + . = ALIGN(L1_CACHE_BYTES); .data.cacheline_aligned : AT(ADDR(.data.cacheline_aligned) - LOAD_OFFSET) { *(.data.cacheline_aligned) } -- cgit v1.2.3 From 8b31e49d1d75729c1da9009664ba52abd1adc628 Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Wed, 27 May 2009 13:50:33 +1000 Subject: powerpc: Fix up dma_alloc_coherent() on platforms without cache coherency. The implementation we just revived has issues, such as using a Kconfig-defined virtual address area in kernel space that nothing actually carves out (and thus will overlap whatever is there), or having some dependencies on being self contained in a single PTE page which adds unnecessary constraints on the kernel virtual address space. This fixes it by using more classic PTE accessors and automatically locating the area for consistent memory, carving an appropriate hole in the kernel virtual address space, leaving only the size of that area as a Kconfig option. It also brings some dma-mask related fixes from the ARM implementation which was almost identical initially but grew its own fixes. Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/dma.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/powerpc/kernel') diff --git a/arch/powerpc/kernel/dma.c b/arch/powerpc/kernel/dma.c index 53c7788cba7..6b02793dc75 100644 --- a/arch/powerpc/kernel/dma.c +++ b/arch/powerpc/kernel/dma.c @@ -32,7 +32,7 @@ void *dma_direct_alloc_coherent(struct device *dev, size_t size, { void *ret; #ifdef CONFIG_NOT_COHERENT_CACHE - ret = __dma_alloc_coherent(size, dma_handle, flag); + ret = __dma_alloc_coherent(dev, size, dma_handle, flag); if (ret == NULL) return NULL; *dma_handle += get_dma_direct_offset(dev); -- cgit v1.2.3