From 51c52e86694f19e84600a40f6156889feafd8ae9 Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Tue, 24 Jun 2008 11:32:36 +1000 Subject: powerpc: Split out do_feature_fixups() from cputable.c The logic to patch CPU feature sections lives in cputable.c, but these days it's used for CPU features as well as firmware features. Move it into it's own file for neatness and as preparation for some additions. While we're moving the code, we pull the loop body logic into a separate routine, and remove a comment which doesn't apply anymore. Signed-off-by: Michael Ellerman Acked-by: Kumar Gala Signed-off-by: Paul Mackerras --- arch/powerpc/lib/feature-fixups.c | 56 +++++++++++++++++++++++++++++++++++++++ 1 file changed, 56 insertions(+) create mode 100644 arch/powerpc/lib/feature-fixups.c (limited to 'arch/powerpc/lib/feature-fixups.c') diff --git a/arch/powerpc/lib/feature-fixups.c b/arch/powerpc/lib/feature-fixups.c new file mode 100644 index 000000000000..f6fd5d2ff10d --- /dev/null +++ b/arch/powerpc/lib/feature-fixups.c @@ -0,0 +1,56 @@ +/* + * Copyright (C) 2001 Ben. Herrenschmidt (benh@kernel.crashing.org) + * + * Modifications for ppc64: + * Copyright (C) 2003 Dave Engebretsen + * + * Copyright 2008 Michael Ellerman, IBM Corporation. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + */ + +#include +#include +#include + + +struct fixup_entry { + unsigned long mask; + unsigned long value; + long start_off; + long end_off; +}; + +static void patch_feature_section(unsigned long value, struct fixup_entry *fcur) +{ + unsigned int *pstart, *pend, *p; + + if ((value & fcur->mask) == fcur->value) + return; + + pstart = ((unsigned int *)fcur) + (fcur->start_off / 4); + pend = ((unsigned int *)fcur) + (fcur->end_off / 4); + + for (p = pstart; p < pend; p++) { + *p = PPC_NOP_INSTR; + asm volatile ("dcbst 0, %0" : : "r" (p)); + } + asm volatile ("sync" : : : "memory"); + for (p = pstart; p < pend; p++) + asm volatile ("icbi 0,%0" : : "r" (p)); + asm volatile ("sync; isync" : : : "memory"); +} + +void do_feature_fixups(unsigned long value, void *fixup_start, void *fixup_end) +{ + struct fixup_entry *fcur, *fend; + + fcur = fixup_start; + fend = fixup_end; + + for (; fcur < fend; fcur++) + patch_feature_section(value, fcur); +} -- cgit v1.2.3 From fac23fe4be23259a8eaa9bad822f5b14dd07d15c Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Tue, 24 Jun 2008 11:32:54 +1000 Subject: powerpc: Introduce infrastructure for feature sections with alternatives The current feature section logic only supports nop'ing out code, this means if you want to choose at runtime between instruction sequences, one or both cases will have to execute the nop'ed out contents of the other section, eg: BEGIN_FTR_SECTION or 1,1,1 END_FTR_SECTION_IFSET(FOO) BEGIN_FTR_SECTION or 2,2,2 END_FTR_SECTION_IFCLR(FOO) and the resulting code will be either, or 1,1,1 nop or, nop or 2,2,2 For small code segments this is fine, but for larger code blocks and in performance criticial code segments, it would be nice to avoid the nops. This commit starts to implement logic to allow the following: BEGIN_FTR_SECTION or 1,1,1 FTR_SECTION_ELSE or 2,2,2 ALT_FTR_SECTION_END_IFSET(FOO) and the resulting code will be: or 1,1,1 or, or 2,2,2 We achieve this by extending the existing FTR macros. The current feature section semantic just becomes a special case, ie. if the else case is empty we nop out the default case. The key limitation is that the size of the else case must be less than or equal to the size of the default case. If the else case is smaller the remainder of the section is nop'ed. We let the linker put the else case code in with the rest of the text, so that relative branches from the else case are more likley to link, this has the disadvantage that we can't free the unused else cases. This commit introduces the required macro and linker script changes, but does not enable the patching of the alternative sections. We also need to update two hand-made section entries in reg.h and timex.h Signed-off-by: Michael Ellerman Signed-off-by: Paul Mackerras --- arch/powerpc/lib/feature-fixups.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'arch/powerpc/lib/feature-fixups.c') diff --git a/arch/powerpc/lib/feature-fixups.c b/arch/powerpc/lib/feature-fixups.c index f6fd5d2ff10d..973d547ef01d 100644 --- a/arch/powerpc/lib/feature-fixups.c +++ b/arch/powerpc/lib/feature-fixups.c @@ -22,6 +22,8 @@ struct fixup_entry { unsigned long value; long start_off; long end_off; + long alt_start_off; + long alt_end_off; }; static void patch_feature_section(unsigned long value, struct fixup_entry *fcur) -- cgit v1.2.3 From 9b1a735de64cc975c31a1642ec55e082ddbdfeaf Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Tue, 24 Jun 2008 11:33:02 +1000 Subject: powerpc: Add logic to patch alternative feature sections This commit adds the logic to patch alternative sections. This is fairly straightforward, except for branches. Relative branches that jump from inside the else section to outside of it need to be translated as they're moved, otherwise they will jump to the wrong location. Signed-off-by: Michael Ellerman Signed-off-by: Paul Mackerras --- arch/powerpc/lib/feature-fixups.c | 79 ++++++++++++++++++++++++++++++++------- 1 file changed, 65 insertions(+), 14 deletions(-) (limited to 'arch/powerpc/lib/feature-fixups.c') diff --git a/arch/powerpc/lib/feature-fixups.c b/arch/powerpc/lib/feature-fixups.c index 973d547ef01d..174c1eba9ac8 100644 --- a/arch/powerpc/lib/feature-fixups.c +++ b/arch/powerpc/lib/feature-fixups.c @@ -26,24 +26,66 @@ struct fixup_entry { long alt_end_off; }; -static void patch_feature_section(unsigned long value, struct fixup_entry *fcur) +static unsigned int *calc_addr(struct fixup_entry *fcur, long offset) { - unsigned int *pstart, *pend, *p; + /* + * We store the offset to the code as a negative offset from + * the start of the alt_entry, to support the VDSO. This + * routine converts that back into an actual address. + */ + return (unsigned int *)((unsigned long)fcur + offset); +} + +static int patch_alt_instruction(unsigned int *src, unsigned int *dest, + unsigned int *alt_start, unsigned int *alt_end) +{ + unsigned int instr; + + instr = *src; + + if (instr_is_relative_branch(*src)) { + unsigned int *target = (unsigned int *)branch_target(src); + + /* Branch within the section doesn't need translating */ + if (target < alt_start || target >= alt_end) { + instr = translate_branch(dest, src); + if (!instr) + return 1; + } + } + + patch_instruction(dest, instr); + + return 0; +} + +static int patch_feature_section(unsigned long value, struct fixup_entry *fcur) +{ + unsigned int *start, *end, *alt_start, *alt_end, *src, *dest; + + start = calc_addr(fcur, fcur->start_off); + end = calc_addr(fcur, fcur->end_off); + alt_start = calc_addr(fcur, fcur->alt_start_off); + alt_end = calc_addr(fcur, fcur->alt_end_off); + + if ((alt_end - alt_start) > (end - start)) + return 1; if ((value & fcur->mask) == fcur->value) - return; + return 0; - pstart = ((unsigned int *)fcur) + (fcur->start_off / 4); - pend = ((unsigned int *)fcur) + (fcur->end_off / 4); + src = alt_start; + dest = start; - for (p = pstart; p < pend; p++) { - *p = PPC_NOP_INSTR; - asm volatile ("dcbst 0, %0" : : "r" (p)); + for (; src < alt_end; src++, dest++) { + if (patch_alt_instruction(src, dest, alt_start, alt_end)) + return 1; } - asm volatile ("sync" : : : "memory"); - for (p = pstart; p < pend; p++) - asm volatile ("icbi 0,%0" : : "r" (p)); - asm volatile ("sync; isync" : : : "memory"); + + for (; dest < end; dest++) + patch_instruction(dest, PPC_NOP_INSTR); + + return 0; } void do_feature_fixups(unsigned long value, void *fixup_start, void *fixup_end) @@ -53,6 +95,15 @@ void do_feature_fixups(unsigned long value, void *fixup_start, void *fixup_end) fcur = fixup_start; fend = fixup_end; - for (; fcur < fend; fcur++) - patch_feature_section(value, fcur); + for (; fcur < fend; fcur++) { + if (patch_feature_section(value, fcur)) { + __WARN(); + printk("Unable to patch feature section at %p - %p" \ + " with %p - %p\n", + calc_addr(fcur, fcur->start_off), + calc_addr(fcur, fcur->end_off), + calc_addr(fcur, fcur->alt_start_off), + calc_addr(fcur, fcur->alt_end_off)); + } + } } -- cgit v1.2.3 From 362e7701fd183d990e0863883461edef61392710 Mon Sep 17 00:00:00 2001 From: Michael Ellerman Date: Tue, 24 Jun 2008 11:33:03 +1000 Subject: powerpc: Add self-tests of the feature fixup code This commit adds tests of the feature fixup code, they are run during boot if CONFIG_FTR_FIXUP_SELFTEST=y. Some of the tests manually invoke the patching routines to check their behaviour, and others use the macros and so are patched during the normal patching done during boot. Because we have two sets of macros with different names, we use a macro to generate the test of the macros, very niiiice. Signed-off-by: Michael Ellerman Signed-off-by: Paul Mackerras --- arch/powerpc/lib/feature-fixups.c | 206 ++++++++++++++++++++++++++++++++++++++ 1 file changed, 206 insertions(+) (limited to 'arch/powerpc/lib/feature-fixups.c') diff --git a/arch/powerpc/lib/feature-fixups.c b/arch/powerpc/lib/feature-fixups.c index 174c1eba9ac8..48e1ed89052d 100644 --- a/arch/powerpc/lib/feature-fixups.c +++ b/arch/powerpc/lib/feature-fixups.c @@ -13,6 +13,8 @@ */ #include +#include +#include #include #include @@ -107,3 +109,207 @@ void do_feature_fixups(unsigned long value, void *fixup_start, void *fixup_end) } } } + +#ifdef CONFIG_FTR_FIXUP_SELFTEST + +#define check(x) \ + if (!(x)) printk("feature-fixups: test failed at line %d\n", __LINE__); + +/* This must be after the text it fixes up, vmlinux.lds.S enforces that atm */ +static struct fixup_entry fixup; + +static long calc_offset(struct fixup_entry *entry, unsigned int *p) +{ + return (unsigned long)p - (unsigned long)entry; +} + +void test_basic_patching(void) +{ + extern unsigned int ftr_fixup_test1; + extern unsigned int end_ftr_fixup_test1; + extern unsigned int ftr_fixup_test1_orig; + extern unsigned int ftr_fixup_test1_expected; + int size = &end_ftr_fixup_test1 - &ftr_fixup_test1; + + fixup.value = fixup.mask = 8; + fixup.start_off = calc_offset(&fixup, &ftr_fixup_test1 + 1); + fixup.end_off = calc_offset(&fixup, &ftr_fixup_test1 + 2); + fixup.alt_start_off = fixup.alt_end_off = 0; + + /* Sanity check */ + check(memcmp(&ftr_fixup_test1, &ftr_fixup_test1_orig, size) == 0); + + /* Check we don't patch if the value matches */ + patch_feature_section(8, &fixup); + check(memcmp(&ftr_fixup_test1, &ftr_fixup_test1_orig, size) == 0); + + /* Check we do patch if the value doesn't match */ + patch_feature_section(0, &fixup); + check(memcmp(&ftr_fixup_test1, &ftr_fixup_test1_expected, size) == 0); + + /* Check we do patch if the mask doesn't match */ + memcpy(&ftr_fixup_test1, &ftr_fixup_test1_orig, size); + check(memcmp(&ftr_fixup_test1, &ftr_fixup_test1_orig, size) == 0); + patch_feature_section(~8, &fixup); + check(memcmp(&ftr_fixup_test1, &ftr_fixup_test1_expected, size) == 0); +} + +static void test_alternative_patching(void) +{ + extern unsigned int ftr_fixup_test2; + extern unsigned int end_ftr_fixup_test2; + extern unsigned int ftr_fixup_test2_orig; + extern unsigned int ftr_fixup_test2_alt; + extern unsigned int ftr_fixup_test2_expected; + int size = &end_ftr_fixup_test2 - &ftr_fixup_test2; + + fixup.value = fixup.mask = 0xF; + fixup.start_off = calc_offset(&fixup, &ftr_fixup_test2 + 1); + fixup.end_off = calc_offset(&fixup, &ftr_fixup_test2 + 2); + fixup.alt_start_off = calc_offset(&fixup, &ftr_fixup_test2_alt); + fixup.alt_end_off = calc_offset(&fixup, &ftr_fixup_test2_alt + 1); + + /* Sanity check */ + check(memcmp(&ftr_fixup_test2, &ftr_fixup_test2_orig, size) == 0); + + /* Check we don't patch if the value matches */ + patch_feature_section(0xF, &fixup); + check(memcmp(&ftr_fixup_test2, &ftr_fixup_test2_orig, size) == 0); + + /* Check we do patch if the value doesn't match */ + patch_feature_section(0, &fixup); + check(memcmp(&ftr_fixup_test2, &ftr_fixup_test2_expected, size) == 0); + + /* Check we do patch if the mask doesn't match */ + memcpy(&ftr_fixup_test2, &ftr_fixup_test2_orig, size); + check(memcmp(&ftr_fixup_test2, &ftr_fixup_test2_orig, size) == 0); + patch_feature_section(~0xF, &fixup); + check(memcmp(&ftr_fixup_test2, &ftr_fixup_test2_expected, size) == 0); +} + +static void test_alternative_case_too_big(void) +{ + extern unsigned int ftr_fixup_test3; + extern unsigned int end_ftr_fixup_test3; + extern unsigned int ftr_fixup_test3_orig; + extern unsigned int ftr_fixup_test3_alt; + int size = &end_ftr_fixup_test3 - &ftr_fixup_test3; + + fixup.value = fixup.mask = 0xC; + fixup.start_off = calc_offset(&fixup, &ftr_fixup_test3 + 1); + fixup.end_off = calc_offset(&fixup, &ftr_fixup_test3 + 2); + fixup.alt_start_off = calc_offset(&fixup, &ftr_fixup_test3_alt); + fixup.alt_end_off = calc_offset(&fixup, &ftr_fixup_test3_alt + 2); + + /* Sanity check */ + check(memcmp(&ftr_fixup_test3, &ftr_fixup_test3_orig, size) == 0); + + /* Expect nothing to be patched, and the error returned to us */ + check(patch_feature_section(0xF, &fixup) == 1); + check(memcmp(&ftr_fixup_test3, &ftr_fixup_test3_orig, size) == 0); + check(patch_feature_section(0, &fixup) == 1); + check(memcmp(&ftr_fixup_test3, &ftr_fixup_test3_orig, size) == 0); + check(patch_feature_section(~0xF, &fixup) == 1); + check(memcmp(&ftr_fixup_test3, &ftr_fixup_test3_orig, size) == 0); +} + +static void test_alternative_case_too_small(void) +{ + extern unsigned int ftr_fixup_test4; + extern unsigned int end_ftr_fixup_test4; + extern unsigned int ftr_fixup_test4_orig; + extern unsigned int ftr_fixup_test4_alt; + extern unsigned int ftr_fixup_test4_expected; + int size = &end_ftr_fixup_test4 - &ftr_fixup_test4; + unsigned long flag; + + /* Check a high-bit flag */ + flag = 1UL << ((sizeof(unsigned long) - 1) * 8); + fixup.value = fixup.mask = flag; + fixup.start_off = calc_offset(&fixup, &ftr_fixup_test4 + 1); + fixup.end_off = calc_offset(&fixup, &ftr_fixup_test4 + 5); + fixup.alt_start_off = calc_offset(&fixup, &ftr_fixup_test4_alt); + fixup.alt_end_off = calc_offset(&fixup, &ftr_fixup_test4_alt + 2); + + /* Sanity check */ + check(memcmp(&ftr_fixup_test4, &ftr_fixup_test4_orig, size) == 0); + + /* Check we don't patch if the value matches */ + patch_feature_section(flag, &fixup); + check(memcmp(&ftr_fixup_test4, &ftr_fixup_test4_orig, size) == 0); + + /* Check we do patch if the value doesn't match */ + patch_feature_section(0, &fixup); + check(memcmp(&ftr_fixup_test4, &ftr_fixup_test4_expected, size) == 0); + + /* Check we do patch if the mask doesn't match */ + memcpy(&ftr_fixup_test4, &ftr_fixup_test4_orig, size); + check(memcmp(&ftr_fixup_test4, &ftr_fixup_test4_orig, size) == 0); + patch_feature_section(~flag, &fixup); + check(memcmp(&ftr_fixup_test4, &ftr_fixup_test4_expected, size) == 0); +} + +static void test_alternative_case_with_branch(void) +{ + extern unsigned int ftr_fixup_test5; + extern unsigned int end_ftr_fixup_test5; + extern unsigned int ftr_fixup_test5_expected; + int size = &end_ftr_fixup_test5 - &ftr_fixup_test5; + + check(memcmp(&ftr_fixup_test5, &ftr_fixup_test5_expected, size) == 0); +} + +static void test_alternative_case_with_external_branch(void) +{ + extern unsigned int ftr_fixup_test6; + extern unsigned int end_ftr_fixup_test6; + extern unsigned int ftr_fixup_test6_expected; + int size = &end_ftr_fixup_test6 - &ftr_fixup_test6; + + check(memcmp(&ftr_fixup_test6, &ftr_fixup_test6_expected, size) == 0); +} + +static void test_cpu_macros(void) +{ + extern void ftr_fixup_test_FTR_macros; + extern void ftr_fixup_test_FTR_macros_expected; + unsigned long size = &ftr_fixup_test_FTR_macros_expected - + &ftr_fixup_test_FTR_macros; + + /* The fixups have already been done for us during boot */ + check(memcmp(&ftr_fixup_test_FTR_macros, + &ftr_fixup_test_FTR_macros_expected, size) == 0); +} + +static void test_fw_macros(void) +{ +#ifdef CONFIG_PPC64 + extern void ftr_fixup_test_FW_FTR_macros; + extern void ftr_fixup_test_FW_FTR_macros_expected; + unsigned long size = &ftr_fixup_test_FW_FTR_macros_expected - + &ftr_fixup_test_FW_FTR_macros; + + /* The fixups have already been done for us during boot */ + check(memcmp(&ftr_fixup_test_FW_FTR_macros, + &ftr_fixup_test_FW_FTR_macros_expected, size) == 0); +#endif +} + +static int __init test_feature_fixups(void) +{ + printk(KERN_DEBUG "Running feature fixup self-tests ...\n"); + + test_basic_patching(); + test_alternative_patching(); + test_alternative_case_too_big(); + test_alternative_case_too_small(); + test_alternative_case_with_branch(); + test_alternative_case_with_external_branch(); + test_cpu_macros(); + test_fw_macros(); + + return 0; +} +late_initcall(test_feature_fixups); + +#endif /* CONFIG_FTR_FIXUP_SELFTEST */ -- cgit v1.2.3 From 2d1b2027626d5151fff8ef7c06ca8e7876a1a510 Mon Sep 17 00:00:00 2001 From: Kumar Gala Date: Wed, 2 Jul 2008 01:16:40 +1000 Subject: powerpc: Fixup lwsync at runtime To allow for a single kernel image on e500 v1/v2/mc we need to fixup lwsync at runtime. On e500v1/v2 lwsync causes an illop so we need to patch up the code. We default to 'sync' since that is always safe and if the cpu is capable we will replace 'sync' with 'lwsync'. We introduce CPU_FTR_LWSYNC as a way to determine at runtime if this is needed. This flag could be moved elsewhere since we dont really use it for the normal CPU_FTR purpose. Finally we only store the relative offset in the fixup section to keep it as small as possible rather than using a full fixup_entry. Signed-off-by: Kumar Gala Signed-off-by: Paul Mackerras --- arch/powerpc/lib/feature-fixups.c | 36 ++++++++++++++++++++++++++++++++++++ 1 file changed, 36 insertions(+) (limited to 'arch/powerpc/lib/feature-fixups.c') diff --git a/arch/powerpc/lib/feature-fixups.c b/arch/powerpc/lib/feature-fixups.c index 48e1ed89052d..4e43702b9813 100644 --- a/arch/powerpc/lib/feature-fixups.c +++ b/arch/powerpc/lib/feature-fixups.c @@ -110,6 +110,22 @@ void do_feature_fixups(unsigned long value, void *fixup_start, void *fixup_end) } } +void do_lwsync_fixups(unsigned long value, void *fixup_start, void *fixup_end) +{ + unsigned int *start, *end, *dest; + + if (!(value & CPU_FTR_LWSYNC)) + return ; + + start = fixup_start; + end = fixup_end; + + for (; start < end; start++) { + dest = (void *)start + *start; + patch_instruction(dest, PPC_LWSYNC_INSTR); + } +} + #ifdef CONFIG_FTR_FIXUP_SELFTEST #define check(x) \ @@ -295,6 +311,25 @@ static void test_fw_macros(void) #endif } +static void test_lwsync_macros(void) +{ + extern void lwsync_fixup_test; + extern void end_lwsync_fixup_test; + extern void lwsync_fixup_test_expected_LWSYNC; + extern void lwsync_fixup_test_expected_SYNC; + unsigned long size = &end_lwsync_fixup_test - + &lwsync_fixup_test; + + /* The fixups have already been done for us during boot */ + if (cur_cpu_spec->cpu_features & CPU_FTR_LWSYNC) { + check(memcmp(&lwsync_fixup_test, + &lwsync_fixup_test_expected_LWSYNC, size) == 0); + } else { + check(memcmp(&lwsync_fixup_test, + &lwsync_fixup_test_expected_SYNC, size) == 0); + } +} + static int __init test_feature_fixups(void) { printk(KERN_DEBUG "Running feature fixup self-tests ...\n"); @@ -307,6 +342,7 @@ static int __init test_feature_fixups(void) test_alternative_case_with_external_branch(); test_cpu_macros(); test_fw_macros(); + test_lwsync_macros(); return 0; } -- cgit v1.2.3