From c75b81bffa0a9493a24ae61d0c52f85e4043c183 Mon Sep 17 00:00:00 2001 From: Nicolas Palix Date: Tue, 24 Aug 2010 17:38:55 +0200 Subject: Coccinelle: Add null/eno.cocci The various basic memory allocation functions don't return ERR_PTR Signed-off-by: Nicolas Palix Signed-off-by: Julia Lawall Signed-off-by: Michal Marek diff --git a/scripts/coccinelle/null/eno.cocci b/scripts/coccinelle/null/eno.cocci new file mode 100644 index 0000000..4c9c52b --- /dev/null +++ b/scripts/coccinelle/null/eno.cocci @@ -0,0 +1,20 @@ +/// The various basic memory allocation functions don't return ERR_PTR +/// +// Confidence: High +// Copyright: (C) 2010 Nicolas Palix, DIKU. GPLv2. +// Copyright: (C) 2010 Julia Lawall, DIKU. GPLv2. +// Copyright: (C) 2010 Gilles Muller, INRIA/LiP6. GPLv2. +// URL: http://coccinelle.lip6.fr/ +// Comments: +// Options: -no_includes -include_headers + +virtual patch + +@@ +expression x,E; +@@ + +x = \(kmalloc\|kzalloc\|kcalloc\|kmem_cache_alloc\|kmem_cache_zalloc\|kmem_cache_alloc_node\|kmalloc_node\|kzalloc_node\)(...) +... when != x = E +- IS_ERR(x) ++ !x -- cgit v0.10.2 From 2e1e1605fc60b3fcf1e9382729a92fc305727952 Mon Sep 17 00:00:00 2001 From: Nicolas Palix Date: Tue, 24 Aug 2010 17:38:56 +0200 Subject: Coccinelle: Add null/kmerr.cocci This semantic patch looks for kmalloc etc that are not followed by a NULL check. It only gives a report in the case where there is some error handling code later in the function, which may be helpful in determining what the error handling code for the call to kmalloc etc should be. Signed-off-by: Nicolas Palix Signed-off-by: Julia Lawall Signed-off-by: Michal Marek diff --git a/scripts/coccinelle/null/kmerr.cocci b/scripts/coccinelle/null/kmerr.cocci new file mode 100644 index 0000000..949bf65 --- /dev/null +++ b/scripts/coccinelle/null/kmerr.cocci @@ -0,0 +1,72 @@ +/// This semantic patch looks for kmalloc etc that are not followed by a +/// NULL check. It only gives a report in the case where there is some +/// error handling code later in the function, which may be helpful +/// in determining what the error handling code for the call to kmalloc etc +/// should be. +/// +// Confidence: High +// Copyright: (C) 2010 Nicolas Palix, DIKU. GPLv2. +// Copyright: (C) 2010 Julia Lawall, DIKU. GPLv2. +// Copyright: (C) 2010 Gilles Muller, INRIA/LiP6. GPLv2. +// URL: http://coccinelle.lip6.fr/ +// Comments: +// Options: -no_includes -include_headers + +virtual context +virtual org +virtual report + +@withtest@ +expression x; +position p; +identifier f,fld; +@@ + +x@p = f(...); +... when != x->fld +\(x == NULL \| x != NULL\) + +@fixed depends on context && !org && !report@ +expression x,x1; +position p1 != withtest.p; +statement S; +position any withtest.p; +identifier f; +@@ + +*x@p1 = \(kmalloc\|kzalloc\|kcalloc\)(...); +... +*x1@p = f(...); +if (!x1) S + +// ------------------------------------------------------------------------ + +@rfixed depends on (org || report) && !context exists@ +expression x,x1; +position p1 != withtest.p; +position p2; +statement S; +position any withtest.p; +identifier f; +@@ + +x@p1 = \(kmalloc\|kzalloc\|kcalloc\)(...); +... +x1@p = f@p2(...); +if (!x1) S + +@script:python depends on org@ +p1 << rfixed.p1; +p2 << rfixed.p2; +@@ + +cocci.print_main("alloc call",p1) +cocci.print_secs("possible model",p2) + +@script:python depends on report@ +p1 << rfixed.p1; +p2 << rfixed.p2; +@@ + +msg = "alloc with no test, possible model on line %s" % (p2[0].line) +coccilib.report.print_report(p1[0],msg) -- cgit v0.10.2 From 7703692ef8d6356c32fbe9665d8cbe5b040eec09 Mon Sep 17 00:00:00 2001 From: Nicolas Palix Date: Tue, 24 Aug 2010 17:38:57 +0200 Subject: Coccinelle: Move deref_null.cocci to a more appropriate directory deref_null.cocci is moved to the 'null' directory which contains other null related rules. Signed-off-by: Nicolas Palix Signed-off-by: Julia Lawall Signed-off-by: Michal Marek diff --git a/scripts/coccinelle/deref_null.cocci b/scripts/coccinelle/deref_null.cocci deleted file mode 100644 index 9969d76..0000000 --- a/scripts/coccinelle/deref_null.cocci +++ /dev/null @@ -1,293 +0,0 @@ -/// -/// A variable is dereference under a NULL test. -/// Even though it is know to be NULL. -/// -// Confidence: Moderate -// Copyright: (C) 2010 Nicolas Palix, DIKU. GPLv2. -// Copyright: (C) 2010 Julia Lawall, DIKU. GPLv2. -// Copyright: (C) 2010 Gilles Muller, INRIA/LiP6. GPLv2. -// URL: http://coccinelle.lip6.fr/ -// Comments: -I ... -all_includes can give more complete results -// Options: - -virtual context -virtual patch -virtual org -virtual report - -@initialize:python depends on !context && patch && !org && !report@ - -import sys -print >> sys.stderr, "This semantic patch does not support the 'patch' mode." - -@depends on patch@ -@@ - -this_rule_should_never_matches(); - -@ifm depends on !patch@ -expression *E; -statement S1,S2; -position p1; -@@ - -if@p1 ((E == NULL && ...) || ...) S1 else S2 - -// The following two rules are separate, because both can match a single -// expression in different ways -@pr1 depends on !patch expression@ -expression *ifm.E; -identifier f; -position p1; -@@ - - (E != NULL && ...) ? <+...E->f@p1...+> : ... - -@pr2 depends on !patch expression@ -expression *ifm.E; -identifier f; -position p2; -@@ - -( - (E != NULL) && ... && <+...E->f@p2...+> -| - (E == NULL) || ... || <+...E->f@p2...+> -| - sizeof(<+...E->f@p2...+>) -) - -// For org and report modes - -@r depends on !context && !patch && (org || report) exists@ -expression subE <= ifm.E; -expression *ifm.E; -expression E1,E2; -identifier f; -statement S1,S2,S3,S4; -iterator iter; -position p!={pr1.p1,pr2.p2}; -position ifm.p1; -@@ - -if@p1 ((E == NULL && ...) || ...) -{ - ... when != if (...) S1 else S2 -( - iter(subE,...) S4 // no use -| - list_remove_head(E2,subE,...) -| - subE = E1 -| - for(subE = E1;...;...) S4 -| - subE++ -| - ++subE -| - --subE -| - subE-- -| - &subE -| - E->f@p // bad use -) - ... when any - return ...; -} -else S3 - -@script:python depends on !context && !patch && !org && report@ -p << r.p; -p1 << ifm.p1; -x << ifm.E; -@@ - -msg="ERROR: %s is NULL but dereferenced." % (x) -coccilib.report.print_report(p[0], msg) -cocci.include_match(False) - -@script:python depends on !context && !patch && org && !report@ -p << r.p; -p1 << ifm.p1; -x << ifm.E; -@@ - -msg="ERROR: %s is NULL but dereferenced." % (x) -msg_safe=msg.replace("[","@(").replace("]",")") -cocci.print_main(msg_safe,p) -cocci.include_match(False) - -@s depends on !context && !patch && (org || report) exists@ -expression subE <= ifm.E; -expression *ifm.E; -expression E1,E2; -identifier f; -statement S1,S2,S3,S4; -iterator iter; -position p!={pr1.p1,pr2.p2}; -position ifm.p1; -@@ - -if@p1 ((E == NULL && ...) || ...) -{ - ... when != if (...) S1 else S2 -( - iter(subE,...) S4 // no use -| - list_remove_head(E2,subE,...) -| - subE = E1 -| - for(subE = E1;...;...) S4 -| - subE++ -| - ++subE -| - --subE -| - subE-- -| - &subE -| - E->f@p // bad use -) - ... when any -} -else S3 - -@script:python depends on !context && !patch && !org && report@ -p << s.p; -p1 << ifm.p1; -x << ifm.E; -@@ - -msg="ERROR: %s is NULL but dereferenced." % (x) -coccilib.report.print_report(p[0], msg) - -@script:python depends on !context && !patch && org && !report@ -p << s.p; -p1 << ifm.p1; -x << ifm.E; -@@ - -msg="ERROR: %s is NULL but dereferenced." % (x) -msg_safe=msg.replace("[","@(").replace("]",")") -cocci.print_main(msg_safe,p) - -// For context mode - -@depends on context && !patch && !org && !report exists@ -expression subE <= ifm.E; -expression *ifm.E; -expression E1,E2; -identifier f; -statement S1,S2,S3,S4; -iterator iter; -position p!={pr1.p1,pr2.p2}; -position ifm.p1; -@@ - -if@p1 ((E == NULL && ...) || ...) -{ - ... when != if (...) S1 else S2 -( - iter(subE,...) S4 // no use -| - list_remove_head(E2,subE,...) -| - subE = E1 -| - for(subE = E1;...;...) S4 -| - subE++ -| - ++subE -| - --subE -| - subE-- -| - &subE -| -* E->f@p // bad use -) - ... when any - return ...; -} -else S3 - -// The following three rules are duplicates of ifm, pr1 and pr2 respectively. -// It is need because the previous rule as already made a "change". - -@ifm1 depends on !patch@ -expression *E; -statement S1,S2; -position p1; -@@ - -if@p1 ((E == NULL && ...) || ...) S1 else S2 - -@pr11 depends on !patch expression@ -expression *ifm1.E; -identifier f; -position p1; -@@ - - (E != NULL && ...) ? <+...E->f@p1...+> : ... - -@pr12 depends on !patch expression@ -expression *ifm1.E; -identifier f; -position p2; -@@ - -( - (E != NULL) && ... && <+...E->f@p2...+> -| - (E == NULL) || ... || <+...E->f@p2...+> -| - sizeof(<+...E->f@p2...+>) -) - -@depends on context && !patch && !org && !report exists@ -expression subE <= ifm1.E; -expression *ifm1.E; -expression E1,E2; -identifier f; -statement S1,S2,S3,S4; -iterator iter; -position p!={pr11.p1,pr12.p2}; -position ifm1.p1; -@@ - -if@p1 ((E == NULL && ...) || ...) -{ - ... when != if (...) S1 else S2 -( - iter(subE,...) S4 // no use -| - list_remove_head(E2,subE,...) -| - subE = E1 -| - for(subE = E1;...;...) S4 -| - subE++ -| - ++subE -| - --subE -| - subE-- -| - &subE -| -* E->f@p // bad use -) - ... when any -} -else S3 diff --git a/scripts/coccinelle/null/deref_null.cocci b/scripts/coccinelle/null/deref_null.cocci new file mode 100644 index 0000000..9969d76 --- /dev/null +++ b/scripts/coccinelle/null/deref_null.cocci @@ -0,0 +1,293 @@ +/// +/// A variable is dereference under a NULL test. +/// Even though it is know to be NULL. +/// +// Confidence: Moderate +// Copyright: (C) 2010 Nicolas Palix, DIKU. GPLv2. +// Copyright: (C) 2010 Julia Lawall, DIKU. GPLv2. +// Copyright: (C) 2010 Gilles Muller, INRIA/LiP6. GPLv2. +// URL: http://coccinelle.lip6.fr/ +// Comments: -I ... -all_includes can give more complete results +// Options: + +virtual context +virtual patch +virtual org +virtual report + +@initialize:python depends on !context && patch && !org && !report@ + +import sys +print >> sys.stderr, "This semantic patch does not support the 'patch' mode." + +@depends on patch@ +@@ + +this_rule_should_never_matches(); + +@ifm depends on !patch@ +expression *E; +statement S1,S2; +position p1; +@@ + +if@p1 ((E == NULL && ...) || ...) S1 else S2 + +// The following two rules are separate, because both can match a single +// expression in different ways +@pr1 depends on !patch expression@ +expression *ifm.E; +identifier f; +position p1; +@@ + + (E != NULL && ...) ? <+...E->f@p1...+> : ... + +@pr2 depends on !patch expression@ +expression *ifm.E; +identifier f; +position p2; +@@ + +( + (E != NULL) && ... && <+...E->f@p2...+> +| + (E == NULL) || ... || <+...E->f@p2...+> +| + sizeof(<+...E->f@p2...+>) +) + +// For org and report modes + +@r depends on !context && !patch && (org || report) exists@ +expression subE <= ifm.E; +expression *ifm.E; +expression E1,E2; +identifier f; +statement S1,S2,S3,S4; +iterator iter; +position p!={pr1.p1,pr2.p2}; +position ifm.p1; +@@ + +if@p1 ((E == NULL && ...) || ...) +{ + ... when != if (...) S1 else S2 +( + iter(subE,...) S4 // no use +| + list_remove_head(E2,subE,...) +| + subE = E1 +| + for(subE = E1;...;...) S4 +| + subE++ +| + ++subE +| + --subE +| + subE-- +| + &subE +| + E->f@p // bad use +) + ... when any + return ...; +} +else S3 + +@script:python depends on !context && !patch && !org && report@ +p << r.p; +p1 << ifm.p1; +x << ifm.E; +@@ + +msg="ERROR: %s is NULL but dereferenced." % (x) +coccilib.report.print_report(p[0], msg) +cocci.include_match(False) + +@script:python depends on !context && !patch && org && !report@ +p << r.p; +p1 << ifm.p1; +x << ifm.E; +@@ + +msg="ERROR: %s is NULL but dereferenced." % (x) +msg_safe=msg.replace("[","@(").replace("]",")") +cocci.print_main(msg_safe,p) +cocci.include_match(False) + +@s depends on !context && !patch && (org || report) exists@ +expression subE <= ifm.E; +expression *ifm.E; +expression E1,E2; +identifier f; +statement S1,S2,S3,S4; +iterator iter; +position p!={pr1.p1,pr2.p2}; +position ifm.p1; +@@ + +if@p1 ((E == NULL && ...) || ...) +{ + ... when != if (...) S1 else S2 +( + iter(subE,...) S4 // no use +| + list_remove_head(E2,subE,...) +| + subE = E1 +| + for(subE = E1;...;...) S4 +| + subE++ +| + ++subE +| + --subE +| + subE-- +| + &subE +| + E->f@p // bad use +) + ... when any +} +else S3 + +@script:python depends on !context && !patch && !org && report@ +p << s.p; +p1 << ifm.p1; +x << ifm.E; +@@ + +msg="ERROR: %s is NULL but dereferenced." % (x) +coccilib.report.print_report(p[0], msg) + +@script:python depends on !context && !patch && org && !report@ +p << s.p; +p1 << ifm.p1; +x << ifm.E; +@@ + +msg="ERROR: %s is NULL but dereferenced." % (x) +msg_safe=msg.replace("[","@(").replace("]",")") +cocci.print_main(msg_safe,p) + +// For context mode + +@depends on context && !patch && !org && !report exists@ +expression subE <= ifm.E; +expression *ifm.E; +expression E1,E2; +identifier f; +statement S1,S2,S3,S4; +iterator iter; +position p!={pr1.p1,pr2.p2}; +position ifm.p1; +@@ + +if@p1 ((E == NULL && ...) || ...) +{ + ... when != if (...) S1 else S2 +( + iter(subE,...) S4 // no use +| + list_remove_head(E2,subE,...) +| + subE = E1 +| + for(subE = E1;...;...) S4 +| + subE++ +| + ++subE +| + --subE +| + subE-- +| + &subE +| +* E->f@p // bad use +) + ... when any + return ...; +} +else S3 + +// The following three rules are duplicates of ifm, pr1 and pr2 respectively. +// It is need because the previous rule as already made a "change". + +@ifm1 depends on !patch@ +expression *E; +statement S1,S2; +position p1; +@@ + +if@p1 ((E == NULL && ...) || ...) S1 else S2 + +@pr11 depends on !patch expression@ +expression *ifm1.E; +identifier f; +position p1; +@@ + + (E != NULL && ...) ? <+...E->f@p1...+> : ... + +@pr12 depends on !patch expression@ +expression *ifm1.E; +identifier f; +position p2; +@@ + +( + (E != NULL) && ... && <+...E->f@p2...+> +| + (E == NULL) || ... || <+...E->f@p2...+> +| + sizeof(<+...E->f@p2...+>) +) + +@depends on context && !patch && !org && !report exists@ +expression subE <= ifm1.E; +expression *ifm1.E; +expression E1,E2; +identifier f; +statement S1,S2,S3,S4; +iterator iter; +position p!={pr11.p1,pr12.p2}; +position ifm1.p1; +@@ + +if@p1 ((E == NULL && ...) || ...) +{ + ... when != if (...) S1 else S2 +( + iter(subE,...) S4 // no use +| + list_remove_head(E2,subE,...) +| + subE = E1 +| + for(subE = E1;...;...) S4 +| + subE++ +| + ++subE +| + --subE +| + subE-- +| + &subE +| +* E->f@p // bad use +) + ... when any +} +else S3 -- cgit v0.10.2 From e105007c1271243568d58c8a4285b19c8eea48d1 Mon Sep 17 00:00:00 2001 From: Nicolas Palix Date: Tue, 24 Aug 2010 17:38:58 +0200 Subject: Coccinelle: Add locks/call_kern.cocci Find functions that refer to GFP_KERNEL but are called with locks held. The proposed change of converting the GFP_KERNEL is not necessarily the correct one. It may be desired to unlock the lock, or to not call the function under the lock in the first place. Signed-off-by: Nicolas Palix Signed-off-by: Julia Lawall Signed-off-by: Michal Marek diff --git a/scripts/coccinelle/locks/call_kern.cocci b/scripts/coccinelle/locks/call_kern.cocci new file mode 100644 index 0000000..00af534 --- /dev/null +++ b/scripts/coccinelle/locks/call_kern.cocci @@ -0,0 +1,74 @@ +/// Find functions that refer to GFP_KERNEL but are called with locks held. +/// The proposed change of converting the GFP_KERNEL is not necessarily the +/// correct one. It may be desired to unlock the lock, or to not call the +/// function under the lock in the first place. +/// +// Confidence: Moderate +// Copyright: (C) 2010 Nicolas Palix, DIKU. GPLv2. +// Copyright: (C) 2010 Julia Lawall, DIKU. GPLv2. +// Copyright: (C) 2010 Gilles Muller, INRIA/LiP6. GPLv2. +// URL: http://coccinelle.lip6.fr/ +// Comments: +// Options: -no_includes -include_headers + +virtual patch + +@gfp exists@ +identifier fn; +position p; +@@ + +fn(...) { + ... when != read_unlock_irq(...) + when != write_unlock_irq(...) + when != read_unlock_irqrestore(...) + when != write_unlock_irqrestore(...) + when != spin_unlock(...) + when != spin_unlock_irq(...) + when != spin_unlock_irqrestore(...) + when != local_irq_enable(...) + when any + GFP_KERNEL@p + ... when any +} + +@locked@ +identifier gfp.fn; +@@ + +( +read_lock_irq +| +write_lock_irq +| +read_lock_irqsave +| +write_lock_irqsave +| +spin_lock +| +spin_trylock +| +spin_lock_irq +| +spin_lock_irqsave +| +local_irq_disable +) + (...) +... when != read_unlock_irq(...) + when != write_unlock_irq(...) + when != read_unlock_irqrestore(...) + when != write_unlock_irqrestore(...) + when != spin_unlock(...) + when != spin_unlock_irq(...) + when != spin_unlock_irqrestore(...) + when != local_irq_enable(...) +fn(...) + +@depends on locked@ +position gfp.p; +@@ + +- GFP_KERNEL@p ++ GFP_ATOMIC -- cgit v0.10.2 From 8aa6273106756c30e5349b58eed28d93d8d88164 Mon Sep 17 00:00:00 2001 From: Nicolas Palix Date: Tue, 24 Aug 2010 17:38:59 +0200 Subject: Coccinelle: Add locks/double_lock.cocci Find double locks. False positives may occur when some paths cannot occur at execution, due to the values of variables, and when there is an intervening function call that releases the lock. Signed-off-by: Nicolas Palix Signed-off-by: Julia Lawall Signed-off-by: Michal Marek diff --git a/scripts/coccinelle/locks/double_lock.cocci b/scripts/coccinelle/locks/double_lock.cocci new file mode 100644 index 0000000..63b24e6 --- /dev/null +++ b/scripts/coccinelle/locks/double_lock.cocci @@ -0,0 +1,92 @@ +/// Find double locks. False positives may occur when some paths cannot +/// occur at execution, due to the values of variables, and when there is +/// an intervening function call that releases the lock. +/// +// Confidence: Moderate +// Copyright: (C) 2010 Nicolas Palix, DIKU. GPLv2. +// Copyright: (C) 2010 Julia Lawall, DIKU. GPLv2. +// Copyright: (C) 2010 Gilles Muller, INRIA/LiP6. GPLv2. +// URL: http://coccinelle.lip6.fr/ +// Comments: +// Options: -no_includes -include_headers + +virtual org +virtual report + +@locked@ +position p1; +expression E1; +position p; +@@ + +( +mutex_lock@p1 +| +mutex_trylock@p1 +| +spin_lock@p1 +| +spin_trylock@p1 +| +read_lock@p1 +| +read_trylock@p1 +| +write_lock@p1 +| +write_trylock@p1 +) (E1@p,...); + +@balanced@ +position p1 != locked.p1; +position locked.p; +identifier lock,unlock; +expression x <= locked.E1; +expression E,locked.E1; +expression E2; +@@ + +if (E) { + <+... when != E1 + lock(E1@p,...) + ...+> +} +... when != E1 + when != \(x = E2\|&x\) + when forall +if (E) { + <+... when != E1 + unlock@p1(E1,...) + ...+> +} + +@r depends on !balanced exists@ +expression x <= locked.E1; +expression locked.E1; +expression E2; +identifier lock; +position locked.p,p1,p2; +@@ + +lock@p1 (E1@p,...); +... when != E1 + when != \(x = E2\|&x\) +lock@p2 (E1,...); + +@script:python depends on org@ +p1 << r.p1; +p2 << r.p2; +lock << r.lock; +@@ + +cocci.print_main(lock,p1) +cocci.print_secs("second lock",p2) + +@script:python depends on report@ +p1 << r.p1; +p2 << r.p2; +lock << r.lock; +@@ + +msg = "second lock on line %s" % (p2[0].line) +coccilib.report.print_report(p1[0],msg) -- cgit v0.10.2 From 54c056280e0dfcd39b33c4ed75116ce98697511c Mon Sep 17 00:00:00 2001 From: Nicolas Palix Date: Tue, 24 Aug 2010 17:39:00 +0200 Subject: Coccinelle: Add locks/mini_lock.cocci Find missing unlocks. This semantic match considers the specific case where the unlock is missing from an if branch, and there is a lock before the if and an unlock after the if. False positives are due to cases where the if branch represents a case where the function is supposed to exit with the lock held, or where there is some preceding function call that releases the lock. Signed-off-by: Nicolas Palix Signed-off-by: Julia Lawall Signed-off-by: Michal Marek diff --git a/scripts/coccinelle/locks/mini_lock.cocci b/scripts/coccinelle/locks/mini_lock.cocci new file mode 100644 index 0000000..7641a29 --- /dev/null +++ b/scripts/coccinelle/locks/mini_lock.cocci @@ -0,0 +1,95 @@ +/// Find missing unlocks. This semantic match considers the specific case +/// where the unlock is missing from an if branch, and there is a lock +/// before the if and an unlock after the if. False positives are due to +/// cases where the if branch represents a case where the function is +/// supposed to exit with the lock held, or where there is some preceding +/// function call that releases the lock. +/// +// Confidence: Moderate +// Copyright: (C) 2010 Nicolas Palix, DIKU. GPLv2. +// Copyright: (C) 2010 Julia Lawall, DIKU. GPLv2. +// Copyright: (C) 2010 Gilles Muller, INRIA/LiP6. GPLv2. +// URL: http://coccinelle.lip6.fr/ +// Comments: +// Options: -no_includes -include_headers + +virtual org +virtual report + +@prelocked@ +position p1,p; +expression E1; +@@ + +( +mutex_lock@p1 +| +mutex_trylock@p1 +| +spin_lock@p1 +| +spin_trylock@p1 +| +read_lock@p1 +| +read_trylock@p1 +| +write_lock@p1 +| +write_trylock@p1 +| +read_lock_irq@p1 +| +write_lock_irq@p1 +| +read_lock_irqsave@p1 +| +write_lock_irqsave@p1 +| +spin_lock_irq@p1 +| +spin_lock_irqsave@p1 +) (E1@p,...); + +@looped@ +position r; +@@ + +for(...;...;...) { <+... return@r ...; ...+> } + +@err@ +expression E1; +position prelocked.p; +position up != prelocked.p1; +position r!=looped.r; +identifier lock,unlock; +@@ + +lock(E1@p,...); +<+... when != E1 +if (...) { + ... when != E1 + return@r ...; +} +...+> +unlock@up(E1,...); + +@script:python depends on org@ +p << prelocked.p1; +lock << err.lock; +unlock << err.unlock; +p2 << err.r; +@@ + +cocci.print_main(lock,p) +cocci.print_secs(unlock,p2) + +@script:python depends on report@ +p << prelocked.p1; +lock << err.lock; +unlock << err.unlock; +p2 << err.r; +@@ + +msg = "preceding lock on line %s" % (p[0].line) +coccilib.report.print_report(p2[0],msg) -- cgit v0.10.2 From 4045159253a49a56d117148d67bc86a72762a0d3 Mon Sep 17 00:00:00 2001 From: Nicolas Palix Date: Tue, 24 Aug 2010 17:39:01 +0200 Subject: Coccinelle: Add iterators/fen.cocci for_each_node iterators only exit normally when the loop cursor is NULL, so there is no point to call of_node_put on the final value. Signed-off-by: Nicolas Palix Signed-off-by: Julia Lawall Signed-off-by: Michal Marek diff --git a/scripts/coccinelle/iterators/fen.cocci b/scripts/coccinelle/iterators/fen.cocci new file mode 100644 index 0000000..77bc108 --- /dev/null +++ b/scripts/coccinelle/iterators/fen.cocci @@ -0,0 +1,64 @@ +/// These iterators only exit normally when the loop cursor is NULL, so there +/// is no point to call of_node_put on the final value. +/// +// Confidence: High +// Copyright: (C) 2010 Nicolas Palix, DIKU. GPLv2. +// Copyright: (C) 2010 Julia Lawall, DIKU. GPLv2. +// Copyright: (C) 2010 Gilles Muller, INRIA/LiP6. GPLv2. +// URL: http://coccinelle.lip6.fr/ +// Comments: +// Options: -no_includes -include_headers + +virtual patch + +@@ +iterator name for_each_node_by_name; +expression np,E; +identifier l; +@@ + +for_each_node_by_name(np,...) { + ... when != break; + when != goto l; +} +... when != np = E +- of_node_put(np); + +@@ +iterator name for_each_node_by_type; +expression np,E; +identifier l; +@@ + +for_each_node_by_type(np,...) { + ... when != break; + when != goto l; +} +... when != np = E +- of_node_put(np); + +@@ +iterator name for_each_compatible_node; +expression np,E; +identifier l; +@@ + +for_each_compatible_node(np,...) { + ... when != break; + when != goto l; +} +... when != np = E +- of_node_put(np); + +@@ +iterator name for_each_matching_node; +expression np,E; +identifier l; +@@ + +for_each_matching_node(np,...) { + ... when != break; + when != goto l; +} +... when != np = E +- of_node_put(np); -- cgit v0.10.2 From afc48a82fafe55dbeeaf7c0d83d29cf279cb1c70 Mon Sep 17 00:00:00 2001 From: Nicolas Palix Date: Tue, 24 Aug 2010 17:39:02 +0200 Subject: Coccinelle: Add iterators/itnull.cocci Many iterators have the property that the first argument is always bound to a real list element, never NULL. False positives arise for some iterators that do not have this property, or in cases when the loop cursor is reassigned. The latter should only happen when the matched code is on the way to a loop exit (break, goto, or return). Signed-off-by: Nicolas Palix Signed-off-by: Julia Lawall Signed-off-by: Michal Marek diff --git a/scripts/coccinelle/iterators/itnull.cocci b/scripts/coccinelle/iterators/itnull.cocci new file mode 100644 index 0000000..baa4297 --- /dev/null +++ b/scripts/coccinelle/iterators/itnull.cocci @@ -0,0 +1,58 @@ +/// Many iterators have the property that the first argument is always bound +/// to a real list element, never NULL. False positives arise for some +/// iterators that do not have this property, or in cases when the loop +/// cursor is reassigned. The latter should only happen when the matched +/// code is on the way to a loop exit (break, goto, or return). +/// +// Confidence: Moderate +// Copyright: (C) 2010 Nicolas Palix, DIKU. GPLv2. +// Copyright: (C) 2010 Julia Lawall, DIKU. GPLv2. +// Copyright: (C) 2010 Gilles Muller, INRIA/LiP6. GPLv2. +// URL: http://coccinelle.lip6.fr/ +// Comments: +// Options: -no_includes -include_headers + +virtual patch + +@@ +iterator I; +expression x,E,E1,E2; +statement S,S1,S2; +@@ + +I(x,...) { <... +( +- if (x == NULL && ...) S +| +- if (x != NULL || ...) + S +| +- (x == NULL) || + E +| +- (x != NULL) && + E +| +- (x == NULL && ...) ? E1 : + E2 +| +- (x != NULL || ...) ? + E1 +- : E2 +| +- if (x == NULL && ...) S1 else + S2 +| +- if (x != NULL || ...) + S1 +- else S2 +| ++ BAD( + x == NULL ++ ) +| ++ BAD( + x != NULL ++ ) +) + ...> } \ No newline at end of file -- cgit v0.10.2 From 5c34050188f5e3db7f8137d819ecbb7d9327dd93 Mon Sep 17 00:00:00 2001 From: Nicolas Palix Date: Tue, 24 Aug 2010 17:39:03 +0200 Subject: Coccinelle: Add iterators/list_entry_update.cocci list_for_each_entry uses its first argument to get from one element of the list to the next, so it is usually not a good idea to reassign it. The first rule finds such a reassignment and the second rule checks that there is a path from the reassignment back to the top of the loop. Signed-off-by: Nicolas Palix Signed-off-by: Julia Lawall Signed-off-by: Michal Marek diff --git a/scripts/coccinelle/iterators/list_entry_update.cocci b/scripts/coccinelle/iterators/list_entry_update.cocci new file mode 100644 index 0000000..b296747 --- /dev/null +++ b/scripts/coccinelle/iterators/list_entry_update.cocci @@ -0,0 +1,62 @@ +/// list_for_each_entry uses its first argument to get from one element of +/// the list to the next, so it is usually not a good idea to reassign it. +/// The first rule finds such a reassignment and the second rule checks +/// that there is a path from the reassignment back to the top of the loop. +/// +// Confidence: High +// Copyright: (C) 2010 Nicolas Palix, DIKU. GPLv2. +// Copyright: (C) 2010 Julia Lawall, DIKU. GPLv2. +// Copyright: (C) 2010 Gilles Muller, INRIA/LiP6. GPLv2. +// URL: http://coccinelle.lip6.fr/ +// Comments: +// Options: -no_includes -include_headers + +virtual context +virtual org +virtual report + +@r@ +iterator name list_for_each_entry; +expression x,E; +position p1,p2; +@@ + +list_for_each_entry@p1(x,...) { <... x =@p2 E ...> } + +@depends on context && !org && !report@ +expression x,E; +position r.p1,r.p2; +statement S; +@@ + +*x =@p2 E +... +list_for_each_entry@p1(x,...) S + +// ------------------------------------------------------------------------ + +@back depends on (org || report) && !context exists@ +expression x,E; +position r.p1,r.p2; +statement S; +@@ + +x =@p2 E +... +list_for_each_entry@p1(x,...) S + +@script:python depends on back && org@ +p1 << r.p1; +p2 << r.p2; +@@ + +cocci.print_main("iterator",p1) +cocci.print_secs("update",p2) + +@script:python depends on back && report@ +p1 << r.p1; +p2 << r.p2; +@@ + +msg = "iterator with update on line %s" % (p2[0].line) +coccilib.report.print_report(p1[0],msg) -- cgit v0.10.2 From 43ba21b57a3e757000bfa8ccf46c81f232b4d881 Mon Sep 17 00:00:00 2001 From: Nicolas Palix Date: Tue, 24 Aug 2010 17:39:04 +0200 Subject: Coccinelle: Add free/kfree.cocci Find a use after free. Values of variables may imply that some execution paths are not possible, resulting in false positives. Another source of false positives are macros such as SCTP_DBG_OBJCNT_DEC that do not actually evaluate their argument Signed-off-by: Nicolas Palix Signed-off-by: Julia Lawall Signed-off-by: Michal Marek diff --git a/scripts/coccinelle/free/kfree.cocci b/scripts/coccinelle/free/kfree.cocci new file mode 100644 index 0000000..c13a539 --- /dev/null +++ b/scripts/coccinelle/free/kfree.cocci @@ -0,0 +1,116 @@ +/// Find a use after free. Values of variables may imply that some +/// execution paths are not possible, resulting in false positives. +/// Another source of false positives are macros such as +/// SCTP_DBG_OBJCNT_DEC that do not actually evaluate their argument +/// +// Confidence: Moderate +// Copyright: (C) 2010 Nicolas Palix, DIKU. GPLv2. +// Copyright: (C) 2010 Julia Lawall, DIKU. GPLv2. +// Copyright: (C) 2010 Gilles Muller, INRIA/LiP6. GPLv2. +// URL: http://coccinelle.lip6.fr/ +// Comments: +// Options: -no_includes -include_headers + +virtual org +virtual report + +@free@ +expression E; +position p1; +@@ + +kfree@p1(E) + +@print expression@ +constant char *c; +expression free.E,E2; +type T; +position p; +identifier f; +@@ + +( + f(...,c,...,(T)E@p,...) +| + E@p == E2 +| + E@p != E2 +| + !E@p +| + E@p || ... +) + +@sz@ +expression free.E; +position p; +@@ + + sizeof(<+...E@p...+>) + +@loop exists@ +expression E; +identifier l; +position ok; +@@ + +while (1) { ... + kfree@ok(E) + ... when != break; + when != goto l; + when forall +} + +@r exists@ +expression free.E, subE<=free.E, E2; +expression E1; +iterator iter; +statement S; +position free.p1!=loop.ok,p2!={print.p,sz.p}; +@@ + +kfree@p1(E,...) +... +( + iter(...,subE,...) S // no use +| + list_remove_head(E1,subE,...) +| + subE = E2 +| + subE++ +| + ++subE +| + --subE +| + subE-- +| + &subE +| + BUG(...) +| + BUG_ON(...) +| + return_VALUE(...) +| + return_ACPI_STATUS(...) +| + E@p2 // bad use +) + +@script:python depends on org@ +p1 << free.p1; +p2 << r.p2; +@@ + +cocci.print_main("kfree",p1) +cocci.print_secs("ref",p2) + +@script:python depends on report@ +p1 << free.p1; +p2 << r.p2; +@@ + +msg = "reference preceded by free on line %s" % (p1[0].line) +coccilib.report.print_report(p2[0],msg) -- cgit v0.10.2 From 998f5523230a8a837809f942d556360f69121677 Mon Sep 17 00:00:00 2001 From: Nicolas Palix Date: Tue, 24 Aug 2010 17:39:05 +0200 Subject: Coccinelle: Add api/kstrdup.cocci Use kstrdup rather than duplicating its implementation Signed-off-by: Nicolas Palix Signed-off-by: Julia Lawall Signed-off-by: Michal Marek diff --git a/scripts/coccinelle/api/kstrdup.cocci b/scripts/coccinelle/api/kstrdup.cocci new file mode 100644 index 0000000..e0805ad --- /dev/null +++ b/scripts/coccinelle/api/kstrdup.cocci @@ -0,0 +1,39 @@ +/// Use kstrdup rather than duplicating its implementation +/// +// Confidence: High +// Copyright: (C) 2010 Nicolas Palix, DIKU. GPLv2. +// Copyright: (C) 2010 Julia Lawall, DIKU. GPLv2. +// Copyright: (C) 2010 Gilles Muller, INRIA/LiP6. GPLv2. +// URL: http://coccinelle.lip6.fr/ +// Comments: +// Options: -no_includes -include_headers + +virtual patch + +@@ +expression from,to; +expression flag,E1,E2; +statement S; +@@ + +- to = kmalloc(strlen(from) + 1,flag); ++ to = kstrdup(from, flag); + ... when != \(from = E1 \| to = E1 \) + if (to==NULL || ...) S + ... when != \(from = E2 \| to = E2 \) +- strcpy(to, from); + +@@ +expression x,from,to; +expression flag,E1,E2,E3; +statement S; +@@ + +- x = strlen(from) + 1; + ... when != \( x = E1 \| from = E1 \) +- to = \(kmalloc\|kzalloc\)(x,flag); ++ to = kstrdup(from, flag); + ... when != \(x = E2 \| from = E2 \| to = E2 \) + if (to==NULL || ...) S + ... when != \(x = E3 \| from = E3 \| to = E3 \) +- memcpy(to, from, x); -- cgit v0.10.2 From 0426a97c4b672166cbcfdc7b18466d4f207f7a36 Mon Sep 17 00:00:00 2001 From: Nicolas Palix Date: Tue, 24 Aug 2010 17:39:06 +0200 Subject: Coccinelle: Add api/memdup.cocci Use kmemdup rather than duplicating its implementation Signed-off-by: Nicolas Palix Signed-off-by: Julia Lawall Signed-off-by: Michal Marek diff --git a/scripts/coccinelle/api/memdup.cocci b/scripts/coccinelle/api/memdup.cocci new file mode 100644 index 0000000..b5d7220 --- /dev/null +++ b/scripts/coccinelle/api/memdup.cocci @@ -0,0 +1,40 @@ +/// Use kmemdup rather than duplicating its implementation +/// +// Confidence: High +// Copyright: (C) 2010 Nicolas Palix, DIKU. GPLv2. +// Copyright: (C) 2010 Julia Lawall, DIKU. GPLv2. +// Copyright: (C) 2010 Gilles Muller, INRIA/LiP6. GPLv2. +// URL: http://coccinelle.lip6.fr/ +// Comments: +// Options: -no_includes -include_headers + +virtual patch + +@r1@ +expression from,to; +expression flag; +position p; +@@ + + to = \(kmalloc@p\|kzalloc@p\)(strlen(from) + 1,flag); + +@r2@ +expression x,from,to; +expression flag,E1; +position p; +@@ + + x = strlen(from) + 1; + ... when != \( x = E1 \| from = E1 \) + to = \(kmalloc@p\|kzalloc@p\)(x,flag); + +@@ +expression from,to,size,flag; +position p != {r1.p,r2.p}; +statement S; +@@ + +- to = \(kmalloc@p\|kzalloc@p\)(size,flag); ++ to = kmemdup(from,size,flag); + if (to==NULL || ...) S +- memcpy(to, from, size); -- cgit v0.10.2 From 034174598bc8ffd9586432852502e368a43ef0f1 Mon Sep 17 00:00:00 2001 From: Nicolas Palix Date: Tue, 24 Aug 2010 17:39:07 +0200 Subject: Coccinelle: Add api/memdup_user Use kmemdup_user rather than duplicating its implementation This is a little bit restricted to reduce false positives Signed-off-by: Nicolas Palix Signed-off-by: Julia Lawall Signed-off-by: Michal Marek diff --git a/scripts/coccinelle/api/memdup_user.cocci b/scripts/coccinelle/api/memdup_user.cocci new file mode 100644 index 0000000..72ce012 --- /dev/null +++ b/scripts/coccinelle/api/memdup_user.cocci @@ -0,0 +1,35 @@ +/// Use kmemdup_user rather than duplicating its implementation +/// This is a little bit restricted to reduce false positives +/// +// Confidence: High +// Copyright: (C) 2010 Nicolas Palix, DIKU. GPLv2. +// Copyright: (C) 2010 Julia Lawall, DIKU. GPLv2. +// Copyright: (C) 2010 Gilles Muller, INRIA/LiP6. GPLv2. +// URL: http://coccinelle.lip6.fr/ +// Comments: +// Options: -no_includes -include_headers + +virtual patch + +@@ +expression from,to,size,flag; +position p; +identifier l1,l2; +@@ + +- to = \(kmalloc@p\|kzalloc@p\)(size,flag); ++ to = memdup_user(from,size); + if ( +- to==NULL ++ IS_ERR(to) + || ...) { + <+... when != goto l1; +- -ENOMEM ++ PTR_ERR(to) + ...+> + } +- if (copy_from_user(to, from, size) != 0) { +- <+... when != goto l2; +- -EFAULT +- ...+> +- } -- cgit v0.10.2 From 99d653d4bc269aebf255f3864334110dbcc70f37 Mon Sep 17 00:00:00 2001 From: Nicolas Palix Date: Tue, 24 Aug 2010 17:39:08 +0200 Subject: Coccinelle: Move err_cast.cocci to the api directory Signed-off-by: Nicolas Palix Signed-off-by: Julia Lawall Signed-off-by: Michal Marek diff --git a/scripts/coccinelle/api/err_cast.cocci b/scripts/coccinelle/api/err_cast.cocci new file mode 100644 index 0000000..2ce11500 --- /dev/null +++ b/scripts/coccinelle/api/err_cast.cocci @@ -0,0 +1,56 @@ +/// +/// Use ERR_CAST inlined function instead of ERR_PTR(PTR_ERR(...)) +/// +// Confidence: High +// Copyright: (C) 2009, 2010 Nicolas Palix, DIKU. GPLv2. +// Copyright: (C) 2009, 2010 Julia Lawall, DIKU. GPLv2. +// Copyright: (C) 2009, 2010 Gilles Muller, INRIA/LiP6. GPLv2. +// URL: http://coccinelle.lip6.fr/ +// Options: +// +// Keywords: ERR_PTR, PTR_ERR, ERR_CAST +// Version min: 2.6.25 +// + +virtual context +virtual patch +virtual org +virtual report + + +@ depends on context && !patch && !org && !report@ +expression x; +@@ + +* ERR_PTR(PTR_ERR(x)) + +@ depends on !context && patch && !org && !report @ +expression x; +@@ + +- ERR_PTR(PTR_ERR(x)) ++ ERR_CAST(x) + +@r depends on !context && !patch && (org || report)@ +expression x; +position p; +@@ + + ERR_PTR@p(PTR_ERR(x)) + +@script:python depends on org@ +p << r.p; +x << r.x; +@@ + +msg="WARNING ERR_CAST can be used with %s" % (x) +msg_safe=msg.replace("[","@(").replace("]",")") +coccilib.org.print_todo(p[0], msg_safe) + +@script:python depends on report@ +p << r.p; +x << r.x; +@@ + +msg="WARNING: ERR_CAST can be used with %s" % (x) +coccilib.report.print_report(p[0], msg) diff --git a/scripts/coccinelle/err_cast.cocci b/scripts/coccinelle/err_cast.cocci deleted file mode 100644 index 2ce11500..0000000 --- a/scripts/coccinelle/err_cast.cocci +++ /dev/null @@ -1,56 +0,0 @@ -/// -/// Use ERR_CAST inlined function instead of ERR_PTR(PTR_ERR(...)) -/// -// Confidence: High -// Copyright: (C) 2009, 2010 Nicolas Palix, DIKU. GPLv2. -// Copyright: (C) 2009, 2010 Julia Lawall, DIKU. GPLv2. -// Copyright: (C) 2009, 2010 Gilles Muller, INRIA/LiP6. GPLv2. -// URL: http://coccinelle.lip6.fr/ -// Options: -// -// Keywords: ERR_PTR, PTR_ERR, ERR_CAST -// Version min: 2.6.25 -// - -virtual context -virtual patch -virtual org -virtual report - - -@ depends on context && !patch && !org && !report@ -expression x; -@@ - -* ERR_PTR(PTR_ERR(x)) - -@ depends on !context && patch && !org && !report @ -expression x; -@@ - -- ERR_PTR(PTR_ERR(x)) -+ ERR_CAST(x) - -@r depends on !context && !patch && (org || report)@ -expression x; -position p; -@@ - - ERR_PTR@p(PTR_ERR(x)) - -@script:python depends on org@ -p << r.p; -x << r.x; -@@ - -msg="WARNING ERR_CAST can be used with %s" % (x) -msg_safe=msg.replace("[","@(").replace("]",")") -coccilib.org.print_todo(p[0], msg_safe) - -@script:python depends on report@ -p << r.p; -x << r.x; -@@ - -msg="WARNING: ERR_CAST can be used with %s" % (x) -coccilib.report.print_report(p[0], msg) -- cgit v0.10.2 From 77c272fb24507c43fa301e3da83eb7d67d36a544 Mon Sep 17 00:00:00 2001 From: Nicolas Palix Date: Tue, 24 Aug 2010 17:39:09 +0200 Subject: Coccinelle: Move resource_size.cocci to the api directory Signed-off-by: Nicolas Palix Signed-off-by: Julia Lawall Signed-off-by: Michal Marek diff --git a/scripts/coccinelle/api/resource_size.cocci b/scripts/coccinelle/api/resource_size.cocci new file mode 100644 index 0000000..1935a58 --- /dev/null +++ b/scripts/coccinelle/api/resource_size.cocci @@ -0,0 +1,93 @@ +/// +/// Use resource_size function on resource object +/// instead of explicit computation. +/// +// Confidence: High +// Copyright: (C) 2009, 2010 Nicolas Palix, DIKU. GPLv2. +// Copyright: (C) 2009, 2010 Julia Lawall, DIKU. GPLv2. +// Copyright: (C) 2009, 2010 Gilles Muller, INRIA/LiP6. GPLv2. +// URL: http://coccinelle.lip6.fr/ +// Options: +// +// Keywords: resource_size +// Version min: 2.6.27 resource_size +// + +virtual context +virtual patch +virtual org +virtual report + +//---------------------------------------------------------- +// For context mode +//---------------------------------------------------------- + +@r_context depends on context && !patch && !org@ +struct resource *res; +@@ + +* (res->end - res->start) + 1 + +//---------------------------------------------------------- +// For patch mode +//---------------------------------------------------------- + +@r_patch depends on !context && patch && !org@ +struct resource *res; +@@ + +- (res->end - res->start) + 1 ++ resource_size(res) + +//---------------------------------------------------------- +// For org mode +//---------------------------------------------------------- + + +@r_org depends on !context && !patch && (org || report)@ +struct resource *res; +position p; +@@ + + (res->end@p - res->start) + 1 + +@rbad_org depends on !context && !patch && (org || report)@ +struct resource *res; +position p != r_org.p; +@@ + + res->end@p - res->start + +@script:python depends on org@ +p << r_org.p; +x << r_org.res; +@@ + +msg="ERROR with %s" % (x) +msg_safe=msg.replace("[","@(").replace("]",")") +coccilib.org.print_todo(p[0], msg_safe) + +@script:python depends on report@ +p << r_org.p; +x << r_org.res; +@@ + +msg="ERROR: Missing resource_size with %s" % (x) +coccilib.report.print_report(p[0], msg) + +@script:python depends on org@ +p << rbad_org.p; +x << rbad_org.res; +@@ + +msg="WARNING with %s" % (x) +msg_safe=msg.replace("[","@(").replace("]",")") +coccilib.org.print_todo(p[0], msg_safe) + +@script:python depends on report@ +p << rbad_org.p; +x << rbad_org.res; +@@ + +msg="WARNING: Suspicious code. resource_size is maybe missing with %s" % (x) +coccilib.report.print_report(p[0], msg) diff --git a/scripts/coccinelle/resource_size.cocci b/scripts/coccinelle/resource_size.cocci deleted file mode 100644 index 1935a58..0000000 --- a/scripts/coccinelle/resource_size.cocci +++ /dev/null @@ -1,93 +0,0 @@ -/// -/// Use resource_size function on resource object -/// instead of explicit computation. -/// -// Confidence: High -// Copyright: (C) 2009, 2010 Nicolas Palix, DIKU. GPLv2. -// Copyright: (C) 2009, 2010 Julia Lawall, DIKU. GPLv2. -// Copyright: (C) 2009, 2010 Gilles Muller, INRIA/LiP6. GPLv2. -// URL: http://coccinelle.lip6.fr/ -// Options: -// -// Keywords: resource_size -// Version min: 2.6.27 resource_size -// - -virtual context -virtual patch -virtual org -virtual report - -//---------------------------------------------------------- -// For context mode -//---------------------------------------------------------- - -@r_context depends on context && !patch && !org@ -struct resource *res; -@@ - -* (res->end - res->start) + 1 - -//---------------------------------------------------------- -// For patch mode -//---------------------------------------------------------- - -@r_patch depends on !context && patch && !org@ -struct resource *res; -@@ - -- (res->end - res->start) + 1 -+ resource_size(res) - -//---------------------------------------------------------- -// For org mode -//---------------------------------------------------------- - - -@r_org depends on !context && !patch && (org || report)@ -struct resource *res; -position p; -@@ - - (res->end@p - res->start) + 1 - -@rbad_org depends on !context && !patch && (org || report)@ -struct resource *res; -position p != r_org.p; -@@ - - res->end@p - res->start - -@script:python depends on org@ -p << r_org.p; -x << r_org.res; -@@ - -msg="ERROR with %s" % (x) -msg_safe=msg.replace("[","@(").replace("]",")") -coccilib.org.print_todo(p[0], msg_safe) - -@script:python depends on report@ -p << r_org.p; -x << r_org.res; -@@ - -msg="ERROR: Missing resource_size with %s" % (x) -coccilib.report.print_report(p[0], msg) - -@script:python depends on org@ -p << rbad_org.p; -x << rbad_org.res; -@@ - -msg="WARNING with %s" % (x) -msg_safe=msg.replace("[","@(").replace("]",")") -coccilib.org.print_todo(p[0], msg_safe) - -@script:python depends on report@ -p << rbad_org.p; -x << rbad_org.res; -@@ - -msg="WARNING: Suspicious code. resource_size is maybe missing with %s" % (x) -coccilib.report.print_report(p[0], msg) -- cgit v0.10.2 From be8fe9d451936a7a412b02cb88a13ffe46879bcf Mon Sep 17 00:00:00 2001 From: Nicolas Palix Date: Tue, 24 Aug 2010 17:39:10 +0200 Subject: Coccinelle: Move alloc directory into api directory alloc contains various semantic patches related to the allocation APIs Signed-off-by: Nicolas Palix Signed-off-by: Julia Lawall Signed-off-by: Michal Marek diff --git a/scripts/coccinelle/alloc/drop_kmalloc_cast.cocci b/scripts/coccinelle/alloc/drop_kmalloc_cast.cocci deleted file mode 100644 index 7d4771d..0000000 --- a/scripts/coccinelle/alloc/drop_kmalloc_cast.cocci +++ /dev/null @@ -1,67 +0,0 @@ -/// -/// Casting (void *) value returned by kmalloc is useless -/// as mentioned in Documentation/CodingStyle, Chap 14. -/// -// Confidence: High -// Copyright: 2009,2010 Nicolas Palix, DIKU. GPLv2. -// URL: http://coccinelle.lip6.fr/ -// Options: -no_includes -include_headers -// -// Keywords: kmalloc, kzalloc, kcalloc -// Version min: < 2.6.12 kmalloc -// Version min: < 2.6.12 kcalloc -// Version min: 2.6.14 kzalloc -// - -virtual context -virtual patch -virtual org -virtual report - -//---------------------------------------------------------- -// For context mode -//---------------------------------------------------------- - -@depends on context@ -type T; -@@ - -* (T *) - \(kmalloc\|kzalloc\|kcalloc\)(...) - -//---------------------------------------------------------- -// For patch mode -//---------------------------------------------------------- - -@depends on patch@ -type T; -@@ - -- (T *) - \(kmalloc\|kzalloc\|kcalloc\)(...) - -//---------------------------------------------------------- -// For org and report mode -//---------------------------------------------------------- - -@r depends on org || report@ -type T; -position p; -@@ - - (T@p *)\(kmalloc\|kzalloc\|kcalloc\)(...) - -@script:python depends on org@ -p << r.p; -t << r.T; -@@ - -coccilib.org.print_safe_todo(p[0], t) - -@script:python depends on report@ -p << r.p; -t << r.T; -@@ - -msg="WARNING: casting value returned by k[cmz]alloc to (%s *) is useless." % (t) -coccilib.report.print_report(p[0], msg) diff --git a/scripts/coccinelle/alloc/kzalloc-simple.cocci b/scripts/coccinelle/alloc/kzalloc-simple.cocci deleted file mode 100644 index 2eae828..0000000 --- a/scripts/coccinelle/alloc/kzalloc-simple.cocci +++ /dev/null @@ -1,82 +0,0 @@ -/// -/// kzalloc should be used rather than kmalloc followed by memset 0 -/// -// Confidence: High -// Copyright: (C) 2009-2010 Julia Lawall, Nicolas Palix, DIKU. GPLv2. -// Copyright: (C) 2009-2010 Gilles Muller, INRIA/LiP6. GPLv2. -// URL: http://coccinelle.lip6.fr/rules/kzalloc.html -// Options: -no_includes -include_headers -// -// Keywords: kmalloc, kzalloc -// Version min: < 2.6.12 kmalloc -// Version min: 2.6.14 kzalloc -// - -virtual context -virtual patch -virtual org -virtual report - -//---------------------------------------------------------- -// For context mode -//---------------------------------------------------------- - -@depends on context@ -type T, T2; -expression x; -expression E1,E2; -statement S; -@@ - -* x = (T)kmalloc(E1,E2); - if ((x==NULL) || ...) S -* memset((T2)x,0,E1); - -//---------------------------------------------------------- -// For patch mode -//---------------------------------------------------------- - -@depends on patch@ -type T, T2; -expression x; -expression E1,E2; -statement S; -@@ - -- x = (T)kmalloc(E1,E2); -+ x = kzalloc(E1,E2); - if ((x==NULL) || ...) S -- memset((T2)x,0,E1); - -//---------------------------------------------------------- -// For org mode -//---------------------------------------------------------- - -@r depends on org || report@ -type T, T2; -expression x; -expression E1,E2; -statement S; -position p; -@@ - - x = (T)kmalloc@p(E1,E2); - if ((x==NULL) || ...) S - memset((T2)x,0,E1); - -@script:python depends on org@ -p << r.p; -x << r.x; -@@ - -msg="%s" % (x) -msg_safe=msg.replace("[","@(").replace("]",")") -coccilib.org.print_todo(p[0], msg_safe) - -@script:python depends on report@ -p << r.p; -x << r.x; -@@ - -msg="WARNING: kzalloc should be used for %s, instead of kmalloc/memset" % (x) -coccilib.report.print_report(p[0], msg) diff --git a/scripts/coccinelle/api/alloc/drop_kmalloc_cast.cocci b/scripts/coccinelle/api/alloc/drop_kmalloc_cast.cocci new file mode 100644 index 0000000..7d4771d --- /dev/null +++ b/scripts/coccinelle/api/alloc/drop_kmalloc_cast.cocci @@ -0,0 +1,67 @@ +/// +/// Casting (void *) value returned by kmalloc is useless +/// as mentioned in Documentation/CodingStyle, Chap 14. +/// +// Confidence: High +// Copyright: 2009,2010 Nicolas Palix, DIKU. GPLv2. +// URL: http://coccinelle.lip6.fr/ +// Options: -no_includes -include_headers +// +// Keywords: kmalloc, kzalloc, kcalloc +// Version min: < 2.6.12 kmalloc +// Version min: < 2.6.12 kcalloc +// Version min: 2.6.14 kzalloc +// + +virtual context +virtual patch +virtual org +virtual report + +//---------------------------------------------------------- +// For context mode +//---------------------------------------------------------- + +@depends on context@ +type T; +@@ + +* (T *) + \(kmalloc\|kzalloc\|kcalloc\)(...) + +//---------------------------------------------------------- +// For patch mode +//---------------------------------------------------------- + +@depends on patch@ +type T; +@@ + +- (T *) + \(kmalloc\|kzalloc\|kcalloc\)(...) + +//---------------------------------------------------------- +// For org and report mode +//---------------------------------------------------------- + +@r depends on org || report@ +type T; +position p; +@@ + + (T@p *)\(kmalloc\|kzalloc\|kcalloc\)(...) + +@script:python depends on org@ +p << r.p; +t << r.T; +@@ + +coccilib.org.print_safe_todo(p[0], t) + +@script:python depends on report@ +p << r.p; +t << r.T; +@@ + +msg="WARNING: casting value returned by k[cmz]alloc to (%s *) is useless." % (t) +coccilib.report.print_report(p[0], msg) diff --git a/scripts/coccinelle/api/alloc/kzalloc-simple.cocci b/scripts/coccinelle/api/alloc/kzalloc-simple.cocci new file mode 100644 index 0000000..046b9b1 --- /dev/null +++ b/scripts/coccinelle/api/alloc/kzalloc-simple.cocci @@ -0,0 +1,86 @@ +/// +/// Use kzalloc rather than kmalloc followed by memset with 0 +/// +/// This considers some simple cases that are common and easy to validate +/// Note in particular that there are no ...s in the rule, so all of the +/// matched code has to be contiguous +/// +// Confidence: High +// Copyright: (C) 2009-2010 Julia Lawall, Nicolas Palix, DIKU. GPLv2. +// Copyright: (C) 2009-2010 Gilles Muller, INRIA/LiP6. GPLv2. +// URL: http://coccinelle.lip6.fr/rules/kzalloc.html +// Options: -no_includes -include_headers +// +// Keywords: kmalloc, kzalloc +// Version min: < 2.6.12 kmalloc +// Version min: 2.6.14 kzalloc +// + +virtual context +virtual patch +virtual org +virtual report + +//---------------------------------------------------------- +// For context mode +//---------------------------------------------------------- + +@depends on context@ +type T, T2; +expression x; +expression E1,E2; +statement S; +@@ + +* x = (T)kmalloc(E1,E2); + if ((x==NULL) || ...) S +* memset((T2)x,0,E1); + +//---------------------------------------------------------- +// For patch mode +//---------------------------------------------------------- + +@depends on patch@ +type T, T2; +expression x; +expression E1,E2; +statement S; +@@ + +- x = (T)kmalloc(E1,E2); ++ x = kzalloc(E1,E2); + if ((x==NULL) || ...) S +- memset((T2)x,0,E1); + +//---------------------------------------------------------- +// For org mode +//---------------------------------------------------------- + +@r depends on org || report@ +type T, T2; +expression x; +expression E1,E2; +statement S; +position p; +@@ + + x = (T)kmalloc@p(E1,E2); + if ((x==NULL) || ...) S + memset((T2)x,0,E1); + +@script:python depends on org@ +p << r.p; +x << r.x; +@@ + +msg="%s" % (x) +msg_safe=msg.replace("[","@(").replace("]",")") +coccilib.org.print_todo(p[0], msg_safe) + +@script:python depends on report@ +p << r.p; +x << r.x; +@@ + +msg="WARNING: kzalloc should be used for %s, instead of kmalloc/memset" % (x) +coccilib.report.print_report(p[0], msg) -- cgit v0.10.2 From 033456e2c507a9b804de56dd258af55f03764334 Mon Sep 17 00:00:00 2001 From: Nicolas Palix Date: Tue, 24 Aug 2010 17:39:11 +0200 Subject: Coccinelle: Add misc/doubleinit.cocci Find duplicate field initializations. This has a high rate of false positives due to #ifdefs, which Coccinelle is not aware of in a structure initialization. Signed-off-by: Nicolas Palix Signed-off-by: Julia Lawall Signed-off-by: Michal Marek diff --git a/scripts/coccinelle/misc/doubleinit.cocci b/scripts/coccinelle/misc/doubleinit.cocci new file mode 100644 index 0000000..55d7dc1 --- /dev/null +++ b/scripts/coccinelle/misc/doubleinit.cocci @@ -0,0 +1,53 @@ +/// Find duplicate field initializations. This has a high rate of false +/// positives due to #ifdefs, which Coccinelle is not aware of in a structure +/// initialization. +/// +// Confidence: Low +// Copyright: (C) 2010 Nicolas Palix, DIKU. GPLv2. +// Copyright: (C) 2010 Julia Lawall, DIKU. GPLv2. +// Copyright: (C) 2010 Gilles Muller, INRIA/LiP6. GPLv2. +// URL: http://coccinelle.lip6.fr/ +// Comments: +// Options: -no_includes -include_headers + +virtual org +virtual report + +@r@ +identifier I, s, fld; +position p0,p; +expression E; +@@ + +struct I s =@p0 { ... .fld@p = E, ...}; + +@s@ +identifier I, s, r.fld; +position r.p0,p; +expression E; +@@ + +struct I s =@p0 { ... .fld@p = E, ...}; + +@script:python depends on org@ +p0 << r.p0; +fld << r.fld; +ps << s.p; +pr << r.p; +@@ + +if int(ps[0].line) < int(pr[0].line) or (int(ps[0].line) == int(pr[0].line) and int(ps[0].column) < int(pr[0].column)): + cocci.print_main(fld,p0) + cocci.print_secs("s",ps) + cocci.print_secs("r",pr) + +@script:python depends on report@ +p0 << r.p0; +fld << r.fld; +ps << s.p; +pr << r.p; +@@ + +if int(ps[0].line) < int(pr[0].line) or (int(ps[0].line) == int(pr[0].line) and int(ps[0].column) < int(pr[0].column)): + msg = "%s: first occurrence %s, second occurrence %s" % (fld,ps[0].line,pr[0].line) + coccilib.report.print_report(p0[0],msg) -- cgit v0.10.2 From 49ab7a39136566eb8312e770031af4182c9f2967 Mon Sep 17 00:00:00 2001 From: Nicolas Palix Date: Tue, 24 Aug 2010 17:39:12 +0200 Subject: Coccinelle: Add misc/ifcol.cocci Find confusingly indented code in or after an if. An if branch should be indented. The code following an if should not be indented. Sometimes, code after an if that is indented is actually intended to be part of the if branch. This has a high rate of false positives, because Coccinelle's column calculation does not distinguish between spaces and tabs, so code that is not visually aligned may be considered to be in the same column. Signed-off-by: Nicolas Palix Signed-off-by: Julia Lawall Signed-off-by: Michal Marek diff --git a/scripts/coccinelle/misc/ifcol.cocci b/scripts/coccinelle/misc/ifcol.cocci new file mode 100644 index 0000000..b7ed91d --- /dev/null +++ b/scripts/coccinelle/misc/ifcol.cocci @@ -0,0 +1,48 @@ +/// Find confusingly indented code in or after an if. An if branch should +/// be indented. The code following an if should not be indented. +/// Sometimes, code after an if that is indented is actually intended to be +/// part of the if branch. +/// +/// This has a high rate of false positives, because Coccinelle's column +/// calculation does not distinguish between spaces and tabs, so code that +/// is not visually aligned may be considered to be in the same column. +/// +// Confidence: Low +// Copyright: (C) 2010 Nicolas Palix, DIKU. GPLv2. +// Copyright: (C) 2010 Julia Lawall, DIKU. GPLv2. +// Copyright: (C) 2010 Gilles Muller, INRIA/LiP6. GPLv2. +// URL: http://coccinelle.lip6.fr/ +// Comments: +// Options: -no_includes -include_headers + +virtual org +virtual report + +@r disable braces4@ +position p1,p2; +statement S1,S2; +@@ + +( +if (...) { ... } +| +if (...) S1@p1 S2@p2 +) + +@script:python depends on org@ +p1 << r.p1; +p2 << r.p2; +@@ + +if (p1[0].column == p2[0].column): + cocci.print_main("branch",p1) + cocci.print_secs("after",p2) + +@script:python depends on report@ +p1 << r.p1; +p2 << r.p2; +@@ + +if (p1[0].column == p2[0].column): + msg = "code aligned with following code on line %s" % (p2[0].line) + coccilib.report.print_report(p1[0],msg) -- cgit v0.10.2 From efdf02cf0651c951de055b0f52128804e11a92fa Mon Sep 17 00:00:00 2001 From: matt mooney Date: Sat, 18 Sep 2010 18:33:57 -0700 Subject: Documentation/kbuild: major edit of modules.txt sections 1-4 Omit needless words and sentences; reorganize and tighten sentence structure; swap sections 2.2 and 2.3 for a more logical flow; remove section 3, therefore shifting 4->3; add to explanations; and add section on building multiple modules. Signed-off-by: matt mooney Signed-off-by: Michal Marek diff --git a/Documentation/kbuild/modules.txt b/Documentation/kbuild/modules.txt index 0767cf6..799b683 100644 --- a/Documentation/kbuild/modules.txt +++ b/Documentation/kbuild/modules.txt @@ -1,215 +1,185 @@ +Building External Modules -In this document you will find information about: -- how to build external modules -- how to make your module use the kbuild infrastructure -- how kbuild will install a kernel -- how to install modules in a non-standard location +This document describes how-to build an out-of-tree kernel module. === Table of Contents === 1 Introduction - === 2 How to build external modules - --- 2.1 Building external modules - --- 2.2 Available targets - --- 2.3 Available options - --- 2.4 Preparing the kernel tree for module build - --- 2.5 Building separate files for a module - === 3. Example commands - === 4. Creating a kbuild file for an external module - === 5. Include files - --- 5.1 How to include files from the kernel include dir - --- 5.2 External modules using an include/ dir - --- 5.3 External modules using several directories - === 6. Module installation - --- 6.1 INSTALL_MOD_PATH - --- 6.2 INSTALL_MOD_DIR - === 7. Module versioning & Module.symvers - --- 7.1 Symbols from the kernel (vmlinux + modules) - --- 7.2 Symbols and external modules - --- 7.3 Symbols from another external module - === 8. Tips & Tricks - --- 8.1 Testing for CONFIG_FOO_BAR + === 2 How-to Build External Modules + --- 2.1 Command Syntax + --- 2.2 Options + --- 2.3 Targets + --- 2.4 Building Separate Files + === 3. Creating a Kbuild File for an External Module + --- 3.1 Shared Makefile + --- 3.2 Separate Kbuild file and Makefile + --- 3.3 Binary Blobs + --- 3.4 Building Multiple Modules + === 4. Include files + --- 4.1 How to include files from the kernel include dir + --- 4.2 External modules using an include/ dir + --- 4.3 External modules using several directories + === 5. Module installation + --- 5.1 INSTALL_MOD_PATH + --- 5.2 INSTALL_MOD_DIR + === 6. Module versioning & Module.symvers + --- 6.1 Symbols from the kernel (vmlinux + modules) + --- 6.2 Symbols and external modules + --- 6.3 Symbols from another external module + === 7. Tips & Tricks + --- 7.1 Testing for CONFIG_FOO_BAR === 1. Introduction -kbuild includes functionality for building modules both -within the kernel source tree and outside the kernel source tree. -The latter is usually referred to as external or "out-of-tree" -modules and is used both during development and for modules that -are not planned to be included in the kernel tree. +"kbuild" is the build system used by the Linux kernel. Modules must use +kbuild to stay compatible with changes in the build infrastructure and +to pick up the right flags to "gcc." Functionality for building modules +both in-tree and out-of-tree is provided. The method for building +either is similar, and all modules are initially developed and built +out-of-tree. -What is covered within this file is mainly information to authors -of modules. The author of an external module should supply -a makefile that hides most of the complexity, so one only has to type -'make' to build the module. A complete example will be presented in -chapter 4, "Creating a kbuild file for an external module". +Covered in this document is information aimed at developers interested +in building out-of-tree (or "external") modules. The author of an +external module should supply a makefile that hides most of the +complexity, so one only has to type "make" to build the module. This is +easily accomplished, and a complete example will be presented in +section 3. -=== 2. How to build external modules +=== 2. How-to Build External Modules -kbuild offers functionality to build external modules, with the -prerequisite that there is a pre-built kernel available with full source. -A subset of the targets available when building the kernel is available -when building an external module. +To build external modules, you must have a pre-built kernel available +that contains the configuration and header files used in the build. +Also, the kernel must have been built with modules enabled. If you are +using a distribution kernel, there will be a package for the kernel you +are running provided by your distribution. ---- 2.1 Building external modules +An alternative is to use the "make" target "modules_prepare." This will +make sure the kernel contains the information required. The target +exists solely as a simple way to prepare a kernel source tree for +building external modules. - Use the following command to build an external module: +NOTE: "modules_prepare" will not build Module.symvers even if +CONFIG_MODVERSIONS is set; therefore, a full kernel build needs to be +executed to make module versioning work. - make -C M=`pwd` +--- 2.1 Command Syntax - For the running kernel use: + The command to build an external module is: - make -C /lib/modules/`uname -r`/build M=`pwd` + make -C M=$PWD - For the above command to succeed, the kernel must have been - built with modules enabled. + The kbuild system knows that an external module is being built + due to the "M=" option given in the command. - To install the modules that were just built: + To build against the running kernel use: - make -C M=`pwd` modules_install + make -C /lib/modules/`uname -r`/build M=$PWD - More complex examples will be shown later, the above should - be enough to get you started. + Then to install the module(s) just built, add the target + "modules_install" to the command: ---- 2.2 Available targets + make -C /lib/modules/`uname -r`/build M=$PWD modules_install - $KDIR refers to the path to the kernel source top-level directory +--- 2.2 Options - make -C $KDIR M=`pwd` - Will build the module(s) located in current directory. - All output files will be located in the same directory - as the module source. - No attempts are made to update the kernel source, and it is - a precondition that a successful make has been executed - for the kernel. + ($KDIR refers to the path of the kernel source directory.) - make -C $KDIR M=`pwd` modules - The modules target is implied when no target is given. - Same functionality as if no target was specified. - See description above. + make -C $KDIR M=$PWD - make -C $KDIR M=`pwd` modules_install - Install the external module(s). - Installation default is in /lib/modules//extra, - but may be prefixed with INSTALL_MOD_PATH - see separate - chapter. + -C $KDIR + The directory where the kernel source is located. + "make" will actually change to the specified directory + when executing and will change back when finished. - make -C $KDIR M=`pwd` clean - Remove all generated files for the module - the kernel - source directory is not modified. + M=$PWD + Informs kbuild that an external module is being built. + The value given to "M" is the absolute path of the + directory where the external module (kbuild file) is + located. - make -C $KDIR M=`pwd` help - help will list the available target when building external - modules. +--- 2.3 Targets ---- 2.3 Available options: + When building an external module, only a subset of the "make" + targets are available. - $KDIR refers to the path to the kernel source top-level directory + make -C $KDIR M=$PWD [target] - make -C $KDIR - Used to specify where to find the kernel source. - '$KDIR' represent the directory where the kernel source is. - Make will actually change directory to the specified directory - when executed but change back when finished. + The default will build the module(s) located in the current + directory, so a target does not need to be specified. All + output files will also be generated in this directory. No + attempts are made to update the kernel source, and it is a + precondition that a successful "make" has been executed for the + kernel. - make -C $KDIR M=`pwd` - M= is used to tell kbuild that an external module is - being built. - The option given to M= is the directory where the external - module (kbuild file) is located. - When an external module is being built only a subset of the - usual targets are available. + modules + The default target for external modules. It has the + same functionality as if no target was specified. See + description above. - make -C $KDIR SUBDIRS=`pwd` - Same as M=. The SUBDIRS= syntax is kept for backwards - compatibility. + modules_install + Install the external module(s). The default location is + /lib/modules//extra, but a prefix may + be added with INSTALL_MOD_PATH (discussed in section 5). ---- 2.4 Preparing the kernel tree for module build + clean + Remove all generated files in the module directory only. - To make sure the kernel contains the information required to - build external modules the target 'modules_prepare' must be used. - 'modules_prepare' exists solely as a simple way to prepare - a kernel source tree for building external modules. - Note: modules_prepare will not build Module.symvers even if - CONFIG_MODVERSIONS is set. Therefore a full kernel build - needs to be executed to make module versioning work. + help + List the available targets for external modules. ---- 2.5 Building separate files for a module - It is possible to build single files which are part of a module. - This works equally well for the kernel, a module and even for - external modules. - Examples (module foo.ko, consist of bar.o, baz.o): - make -C $KDIR M=`pwd` bar.lst - make -C $KDIR M=`pwd` bar.o - make -C $KDIR M=`pwd` foo.ko - make -C $KDIR M=`pwd` / - - -=== 3. Example commands - -This example shows the actual commands to be executed when building -an external module for the currently running kernel. -In the example below, the distribution is supposed to use the -facility to locate output files for a kernel compile in a different -directory than the kernel source - but the examples will also work -when the source and the output files are mixed in the same directory. - -# Kernel source -/lib/modules//source -> /usr/src/linux- +--- 2.4 Building Separate Files -# Output from kernel compile -/lib/modules//build -> /usr/src/linux--up - -Change to the directory where the kbuild file is located and execute -the following commands to build the module: + It is possible to build single files that are part of a module. + This works equally well for the kernel, a module, and even for + external modules. - cd /home/user/src/module - make -C /usr/src/`uname -r`/source \ - O=/lib/modules/`uname-r`/build \ - M=`pwd` + Example (The module foo.ko, consist of bar.o and baz.o): + make -C $KDIR M=$PWD bar.lst + make -C $KDIR M=$PWD baz.o + make -C $KDIR M=$PWD foo.ko + make -C $KDIR M=$PWD / -Then, to install the module use the following command: - make -C /usr/src/`uname -r`/source \ - O=/lib/modules/`uname-r`/build \ - M=`pwd` \ - modules_install +=== 3. Creating a Kbuild File for an External Module -If you look closely you will see that this is the same command as -listed before - with the directories spelled out. +In the last section we saw the command to build a module for the +running kernel. The module is not actually built, however, because a +build file is required. Contained in this file will be the name of +the module(s) being built, along with the list of requisite source +files. The file may be as simple as a single line: -The above are rather long commands, and the following chapter -lists a few tricks to make it all easier. + obj-m := .o +The kbuild system will build .o from .c, +and, after linking, will result in the kernel module .ko. +The above line can be put in either a "Kbuild" file or a "Makefile." +When the module is built from multiple sources, an additional line is +needed listing the files: -=== 4. Creating a kbuild file for an external module + -y := .o .o ... -kbuild is the build system for the kernel, and external modules -must use kbuild to stay compatible with changes in the build system -and to pick up the right flags to gcc etc. +NOTE: Further documentation describing the syntax used by kbuild is +located in Documentation/kbuild/makefiles.txt. -The kbuild file used as input shall follow the syntax described -in Documentation/kbuild/makefiles.txt. This chapter will introduce a few -more tricks to be used when dealing with external modules. +The examples below demonstrate how-to create a build file for the +module 8123.ko, which is built from the following files: -In the following a Makefile will be created for a module with the -following files: 8123_if.c 8123_if.h 8123_pci.c 8123_bin.o_shipped <= Binary blob ---- 4.1 Shared Makefile for module and kernel +--- 3.1 Shared Makefile - An external module always includes a wrapper Makefile supporting - building the module using 'make' with no arguments. - The Makefile provided will most likely include additional - functionality such as test targets etc. and this part shall - be filtered away from kbuild since it may impact kbuild if - name clashes occurs. + An external module always includes a wrapper makefile that + supports building the module using "make" with no arguments. + This target is not used by kbuild; it is only for convenience. + Additional functionality, such as test targets, can be included + but should be filtered out from kbuild due to possible name + clashes. Example 1: --> filename: Makefile @@ -219,11 +189,11 @@ following files: 8123-y := 8123_if.o 8123_pci.o 8123_bin.o else - # Normal Makefile + # normal makefile + KDIR ?= /lib/modules/`uname -r`/build - KERNELDIR := /lib/modules/`uname -r`/build - all:: - $(MAKE) -C $(KERNELDIR) M=`pwd` $@ + default: + $(MAKE) -C $(KDIR) M=$$PWD # Module specific targets genbin: @@ -231,15 +201,20 @@ following files: endif - In example 1, the check for KERNELRELEASE is used to separate - the two parts of the Makefile. kbuild will only see the two - assignments whereas make will see everything except the two - kbuild assignments. + The check for KERNELRELEASE is used to separate the two parts + of the makefile. In the example, kbuild will only see the two + assignments, whereas "make" will see everything except these + two assignments. This is due to two passes made on the file: + the first pass is by the "make" instance run on the + command line; the second pass is by the kbuild system, which is + initiated by the parameterized "make" in the default target. - In recent versions of the kernel, kbuild will look for a file named - Kbuild and as second option look for a file named Makefile. - Utilising the Kbuild file makes us split up the Makefile in example 1 - into two files as shown in example 2: +--- 3.2 Separate Kbuild File and Makefile + + In newer versions of the kernel, kbuild will first look for a + file named "Kbuild", and only if that is not found, will it + then look for a makefile. Utilizing a "Kbuild" file allows us + to split up the makefile from example 1 into two files: Example 2: --> filename: Kbuild @@ -247,20 +222,21 @@ following files: 8123-y := 8123_if.o 8123_pci.o 8123_bin.o --> filename: Makefile - KERNELDIR := /lib/modules/`uname -r`/build - all:: - $(MAKE) -C $(KERNELDIR) M=`pwd` $@ + KDIR ?= /lib/modules/`uname -r`/build + + default: + $(MAKE) -C $(KDIR) M=$$PWD # Module specific targets genbin: echo "X" > 8123_bin.o_shipped + The split in example 2 is questionable due to the simplicity of + each file; however, some external modules use makefiles + consisting of several hundred lines, and here it really pays + off to separate the kbuild part from the rest. - In example 2, we are down to two fairly simple files and for simple - files as used in this example the split is questionable. But some - external modules use Makefiles of several hundred lines and here it - really pays off to separate the kbuild part from the rest. - Example 3 shows a backward compatible version. + The next example shows a backward compatible version. Example 3: --> filename: Kbuild @@ -269,13 +245,15 @@ following files: --> filename: Makefile ifneq ($(KERNELRELEASE),) + # kbuild part of makefile include Kbuild + else - # Normal Makefile + # normal makefile + KDIR ?= /lib/modules/`uname -r`/build - KERNELDIR := /lib/modules/`uname -r`/build - all:: - $(MAKE) -C $(KERNELDIR) M=`pwd` $@ + default: + $(MAKE) -C $(KDIR) M=$$PWD # Module specific targets genbin: @@ -283,28 +261,41 @@ following files: endif - The trick here is to include the Kbuild file from Makefile, so - if an older version of kbuild picks up the Makefile, the Kbuild - file will be included. + Here the "Kbuild" file is included from the makefile. This + allows an older version of kbuild, which only knows of + makefiles, to be used when the "make" and kbuild parts are + split into separate files. ---- 4.2 Binary blobs included in a module +--- 3.3 Binary Blobs - Some external modules needs to include a .o as a blob. kbuild - has support for this, but requires the blob file to be named - _shipped. In our example the blob is named - 8123_bin.o_shipped and when the kbuild rules kick in the file - 8123_bin.o is created as a simple copy off the 8213_bin.o_shipped file - with the _shipped part stripped of the filename. - This allows the 8123_bin.o filename to be used in the assignment to - the module. + Some external modules need to include an object file as a blob. + kbuild has support for this, but requires the blob file to be + named _shipped. When the kbuild rules kick in, a copy + of _shipped is created with _shipped stripped off, + giving us . This shortened filename can be used in + the assignment to the module. + + Throughout this section, 8123_bin.o_shipped has been used to + build the kernel module 8123.ko; it has been included as + 8123_bin.o. - Example 4: - obj-m := 8123.o 8123-y := 8123_if.o 8123_pci.o 8123_bin.o - In example 4, there is no distinction between the ordinary .c/.h files - and the binary file. But kbuild will pick up different rules to create - the .o file. + Although there is no distinction between the ordinary source + files and the binary file, kbuild will pick up different rules + when creating the object file for the module. + +--- 3.4 Building Multiple Modules + + kbuild supports building multiple modules with a single build + file. For example, if you want to build two modules, foo and + bar, the kbuild lines would be: + + obj-m := foo.o bar.o + foo-y := + bar-y := + + It is that simple! === 5. Include files -- cgit v0.10.2 From 9f02186c236ba4ce11ba4e50d5a756c53d56c29e Mon Sep 17 00:00:00 2001 From: matt mooney Date: Sun, 19 Sep 2010 23:06:36 -0700 Subject: Documentation/kbuild: major edit of modules.txt sections 5-8 A follow-up to my edit of the first 4 sections. Shift sections down by one due to the deletion of section 3; grammar corrections along with some rewording; margin width cleanup; and change EXTRA_CFLAGS -> ccflags-y. Signed-off-by: matt mooney Signed-off-by: Michal Marek diff --git a/Documentation/kbuild/modules.txt b/Documentation/kbuild/modules.txt index 799b683..b572db3 100644 --- a/Documentation/kbuild/modules.txt +++ b/Documentation/kbuild/modules.txt @@ -15,17 +15,17 @@ This document describes how-to build an out-of-tree kernel module. --- 3.2 Separate Kbuild file and Makefile --- 3.3 Binary Blobs --- 3.4 Building Multiple Modules - === 4. Include files - --- 4.1 How to include files from the kernel include dir - --- 4.2 External modules using an include/ dir - --- 4.3 External modules using several directories - === 5. Module installation + === 4. Include Files + --- 4.1 Kernel Includes + --- 4.2 Single Subdirectory + --- 4.3 Several Subdirectories + === 5. Module Installation --- 5.1 INSTALL_MOD_PATH --- 5.2 INSTALL_MOD_DIR - === 6. Module versioning & Module.symvers - --- 6.1 Symbols from the kernel (vmlinux + modules) - --- 6.2 Symbols and external modules - --- 6.3 Symbols from another external module + === 6. Module Versioning + --- 6.1 Symbols From the Kernel (vmlinux + modules) + --- 6.2 Symbols and External Modules + --- 6.3 Symbols From Another External Module === 7. Tips & Tricks --- 7.1 Testing for CONFIG_FOO_BAR @@ -298,236 +298,232 @@ module 8123.ko, which is built from the following files: It is that simple! -=== 5. Include files +=== 4. Include Files -Include files are a necessity when a .c file uses something from other .c -files (not strictly in the sense of C, but if good programming practice is -used). Any module that consists of more than one .c file will have a .h file -for one of the .c files. +Within the kernel, header files are kept in standard locations +according to the following rule: -- If the .h file only describes a module internal interface, then the .h file - shall be placed in the same directory as the .c files. -- If the .h files describe an interface used by other parts of the kernel - located in different directories, the .h files shall be located in - include/linux/ or other include/ directories as appropriate. + * If the header file only describes the internal interface of a + module, then the file is placed in the same directory as the + source files. + * If the header file describes an interface used by other parts + of the kernel that are located in different directories, then + the file is placed in include/linux/. -One exception for this rule is larger subsystems that have their own directory -under include/ such as include/scsi. Another exception is arch-specific -.h files which are located under include/asm-$(ARCH)/*. + NOTE: There are two notable exceptions to this rule: larger + subsystems have their own directory under include/, such as + include/scsi; and architecture specific headers are located + under arch/$(ARCH)/include/. -External modules have a tendency to locate include files in a separate include/ -directory and therefore need to deal with this in their kbuild file. +--- 4.1 Kernel Includes ---- 5.1 How to include files from the kernel include dir - - When a module needs to include a file from include/linux/, then one - just uses: + To include a header file located under include/linux/, simply + use: #include - kbuild will make sure to add options to gcc so the relevant - directories are searched. - Likewise for .h files placed in the same directory as the .c file. - - #include "8123_if.h" - - will do the job. + kbuild will add options to "gcc" so the relevant directories + are searched. ---- 5.2 External modules using an include/ dir +--- 4.2 Single Subdirectory - External modules often locate their .h files in a separate include/ - directory although this is not usual kernel style. When an external - module uses an include/ dir then kbuild needs to be told so. - The trick here is to use either EXTRA_CFLAGS (take effect for all .c - files) or CFLAGS_$F.o (take effect only for a single file). + External modules tend to place header files in a separate + include/ directory where their source is located, although this + is not the usual kernel style. To inform kbuild of the + directory use either ccflags-y or CFLAGS_.o. - In our example, if we move 8123_if.h to a subdirectory named include/ - the resulting Kbuild file would look like: + Using the example from section 3, if we moved 8123_if.h to a + subdirectory named include, the resulting kbuild file would + look like: --> filename: Kbuild - obj-m := 8123.o + obj-m := 8123.o - EXTRA_CFLAGS := -Iinclude + ccflags-y := -Iinclude 8123-y := 8123_if.o 8123_pci.o 8123_bin.o - Note that in the assignment there is no space between -I and the path. - This is a kbuild limitation: there must be no space present. - ---- 5.3 External modules using several directories + Note that in the assignment there is no space between -I and + the path. This is a limitation of kbuild: there must be no + space present. - If an external module does not follow the usual kernel style, but - decides to spread files over several directories, then kbuild can - handle this too. +--- 4.3 Several Subdirectories + kbuild can handle files that are spread over several directories. Consider the following example: - | - +- src/complex_main.c - | +- hal/hardwareif.c - | +- hal/include/hardwareif.h - +- include/complex.h - - To build a single module named complex.ko, we then need the following + . + |__ src + | |__ complex_main.c + | |__ hal + | |__ hardwareif.c + | |__ include + | |__ hardwareif.h + |__ include + |__ complex.h + + To build the module complex.ko, we then need the following kbuild file: - Kbuild: + --> filename: Kbuild obj-m := complex.o complex-y := src/complex_main.o complex-y += src/hal/hardwareif.o - EXTRA_CFLAGS := -I$(src)/include - EXTRA_CFLAGS += -I$(src)src/hal/include + ccflags-y := -I$(src)/include + ccflags-y += -I$(src)/src/hal/include + As you can see, kbuild knows how to handle object files located + in other directories. The trick is to specify the directory + relative to the kbuild file's location. That being said, this + is NOT recommended practice. - kbuild knows how to handle .o files located in another directory - - although this is NOT recommended practice. The syntax is to specify - the directory relative to the directory where the Kbuild file is - located. + For the header files, kbuild must be explicitly told where to + look. When kbuild executes, the current directory is always the + root of the kernel tree (the argument to "-C") and therefore an + absolute path is needed. $(src) provides the absolute path by + pointing to the directory where the currently executing kbuild + file is located. - To find the .h files, we have to explicitly tell kbuild where to look - for the .h files. When kbuild executes, the current directory is always - the root of the kernel tree (argument to -C) and therefore we have to - tell kbuild how to find the .h files using absolute paths. - $(src) will specify the absolute path to the directory where the - Kbuild file are located when being build as an external module. - Therefore -I$(src)/ is used to point out the directory of the Kbuild - file and any additional path are just appended. -=== 6. Module installation +=== 5. Module Installation -Modules which are included in the kernel are installed in the directory: +Modules which are included in the kernel are installed in the +directory: /lib/modules/$(KERNELRELEASE)/kernel -External modules are installed in the directory: +And external modules are installed in: /lib/modules/$(KERNELRELEASE)/extra ---- 6.1 INSTALL_MOD_PATH +--- 5.1 INSTALL_MOD_PATH - Above are the default directories, but as always, some level of - customization is possible. One can prefix the path using the variable - INSTALL_MOD_PATH: + Above are the default directories but as always some level of + customization is possible. A prefix can be added to the + installation path using the variable INSTALL_MOD_PATH: $ make INSTALL_MOD_PATH=/frodo modules_install => Install dir: /frodo/lib/modules/$(KERNELRELEASE)/kernel - INSTALL_MOD_PATH may be set as an ordinary shell variable or as in the - example above, can be specified on the command line when calling make. - INSTALL_MOD_PATH has effect both when installing modules included in - the kernel as well as when installing external modules. + INSTALL_MOD_PATH may be set as an ordinary shell variable or, + as shown above, can be specified on the command line when + calling "make." This has effect when installing both in-tree + and out-of-tree modules. ---- 6.2 INSTALL_MOD_DIR +--- 5.2 INSTALL_MOD_DIR - When installing external modules they are by default installed to a - directory under /lib/modules/$(KERNELRELEASE)/extra, but one may wish - to locate modules for a specific functionality in a separate - directory. For this purpose, one can use INSTALL_MOD_DIR to specify an - alternative name to 'extra'. + External modules are by default installed to a directory under + /lib/modules/$(KERNELRELEASE)/extra, but you may wish to locate + modules for a specific functionality in a separate directory. + For this purpose, use INSTALL_MOD_DIR to specify an alternative + name to "extra." - $ make INSTALL_MOD_DIR=gandalf -C KERNELDIR \ - M=`pwd` modules_install + $ make INSTALL_MOD_DIR=gandalf -C $KDIR \ + M=$PWD modules_install => Install dir: /lib/modules/$(KERNELRELEASE)/gandalf -=== 7. Module versioning & Module.symvers +=== 6. Module Versioning -Module versioning is enabled by the CONFIG_MODVERSIONS tag. +Module versioning is enabled by the CONFIG_MODVERSIONS tag, and is used +as a simple ABI consistency check. A CRC value of the full prototype +for an exported symbol is created. When a module is loaded/used, the +CRC values contained in the kernel are compared with similar values in +the module; if they are not equal, the kernel refuses to load the +module. -Module versioning is used as a simple ABI consistency check. The Module -versioning creates a CRC value of the full prototype for an exported symbol and -when a module is loaded/used then the CRC values contained in the kernel are -compared with similar values in the module. If they are not equal, then the -kernel refuses to load the module. +Module.symvers contains a list of all exported symbols from a kernel +build. -Module.symvers contains a list of all exported symbols from a kernel build. +--- 6.1 Symbols From the Kernel (vmlinux + modules) ---- 7.1 Symbols from the kernel (vmlinux + modules) - - During a kernel build, a file named Module.symvers will be generated. - Module.symvers contains all exported symbols from the kernel and - compiled modules. For each symbols, the corresponding CRC value - is stored too. + During a kernel build, a file named Module.symvers will be + generated. Module.symvers contains all exported symbols from + the kernel and compiled modules. For each symbol, the + corresponding CRC value is also stored. The syntax of the Module.symvers file is: - - Sample: + + 0x2d036834 scsi_remove_host drivers/scsi/scsi_mod - For a kernel build without CONFIG_MODVERSIONS enabled, the crc - would read: 0x00000000 + For a kernel build without CONFIG_MODVERSIONS enabled, the CRC + would read 0x00000000. Module.symvers serves two purposes: - 1) It lists all exported symbols both from vmlinux and all modules - 2) It lists the CRC if CONFIG_MODVERSIONS is enabled - ---- 7.2 Symbols and external modules - - When building an external module, the build system needs access to - the symbols from the kernel to check if all external symbols are - defined. This is done in the MODPOST step and to obtain all - symbols, modpost reads Module.symvers from the kernel. - If a Module.symvers file is present in the directory where - the external module is being built, this file will be read too. - During the MODPOST step, a new Module.symvers file will be written - containing all exported symbols that were not defined in the kernel. - ---- 7.3 Symbols from another external module - - Sometimes, an external module uses exported symbols from another - external module. Kbuild needs to have full knowledge on all symbols - to avoid spitting out warnings about undefined symbols. - Three solutions exist to let kbuild know all symbols of more than - one external module. - The method with a top-level kbuild file is recommended but may be - impractical in certain situations. - - Use a top-level Kbuild file - If you have two modules: 'foo' and 'bar', and 'foo' needs - symbols from 'bar', then one can use a common top-level kbuild - file so both modules are compiled in same build. - - Consider following directory layout: - ./foo/ <= contains the foo module - ./bar/ <= contains the bar module - The top-level Kbuild file would then look like: - - #./Kbuild: (this file may also be named Makefile) + 1) It lists all exported symbols from vmlinux and all modules. + 2) It lists the CRC if CONFIG_MODVERSIONS is enabled. + +--- 6.2 Symbols and External Modules + + When building an external module, the build system needs access + to the symbols from the kernel to check if all external symbols + are defined. This is done in the MODPOST step. modpost obtains + the symbols by reading Module.symvers from the kernel source + tree. If a Module.symvers file is present in the directory + where the external module is being built, this file will be + read too. During the MODPOST step, a new Module.symvers file + will be written containing all exported symbols that were not + defined in the kernel. + +--- 6.3 Symbols From Another External Module + + Sometimes, an external module uses exported symbols from + another external module. kbuild needs to have full knowledge of + all symbols to avoid spitting out warnings about undefined + symbols. Three solutions exist for this situation. + + NOTE: The method with a top-level kbuild file is recommended + but may be impractical in certain situations. + + Use a top-level kbuild file + If you have two modules, foo.ko and bar.ko, where + foo.ko needs symbols from bar.ko, then you can use a + common top-level kbuild file so both modules are + compiled in the same build. Consider following + directory layout: + + ./foo/ <= contains foo.ko + ./bar/ <= contains bar.ko + + The top-level kbuild file would then look like: + + #./Kbuild (or ./Makefile): obj-y := foo/ bar/ - Executing: - make -C $KDIR M=`pwd` + And executing: + $ make -C $KDIR M=$PWD - will then do the expected and compile both modules with full - knowledge on symbols from both modules. + Will then do the expected and compile both modules with + full knowledge of symbols from either module. Use an extra Module.symvers file - When an external module is built, a Module.symvers file is - generated containing all exported symbols which are not - defined in the kernel. - To get access to symbols from module 'bar', one can copy the - Module.symvers file from the compilation of the 'bar' module - to the directory where the 'foo' module is built. - During the module build, kbuild will read the Module.symvers - file in the directory of the external module and when the - build is finished, a new Module.symvers file is created - containing the sum of all symbols defined and not part of the - kernel. - - Use make variable KBUILD_EXTRA_SYMBOLS in the Makefile - If it is impractical to copy Module.symvers from another - module, you can assign a space separated list of files to - KBUILD_EXTRA_SYMBOLS in your Makfile. These files will be - loaded by modpost during the initialisation of its symbol - tables. - -=== 8. Tips & Tricks - ---- 8.1 Testing for CONFIG_FOO_BAR - - Modules often need to check for certain CONFIG_ options to decide if - a specific feature shall be included in the module. When kbuild is used - this is done by referencing the CONFIG_ variable directly. + When an external module is built, a Module.symvers file + is generated containing all exported symbols which are + not defined in the kernel. To get access to symbols + from bar.ko, copy the Module.symvers file from the + compilation of bar.ko to the directory where foo.ko is + built. During the module build, kbuild will read the + Module.symvers file in the directory of the external + module, and when the build is finished, a new + Module.symvers file is created containing the sum of + all symbols defined and not part of the kernel. + + Use "make" variable KBUILD_EXTRA_SYMBOLS + If it is impractical to copy Module.symvers from + another module, you can assign a space separated list + of files to KBUILD_EXTRA_SYMBOLS in your build + file. These files will be loaded by modpost during the + initialization of its symbol tables. + +=== 7. Tips & Tricks + +--- 7.1 Testing for CONFIG_FOO_BAR + + Modules often need to check for certain CONFIG_ options to + decide if a specific feature is included in the module. In + kbuild this is done by referencing the CONFIG_ variable + directly. #fs/ext2/Makefile obj-$(CONFIG_EXT2_FS) += ext2.o @@ -535,9 +531,9 @@ Module.symvers contains a list of all exported symbols from a kernel build. ext2-y := balloc.o bitmap.o dir.o ext2-$(CONFIG_EXT2_FS_XATTR) += xattr.o - External modules have traditionally used grep to check for specific - CONFIG_ settings directly in .config. This usage is broken. - As introduced before, external modules shall use kbuild when building - and therefore can use the same methods as in-kernel modules when - testing for CONFIG_ definitions. + External modules have traditionally used "grep" to check for + specific CONFIG_ settings directly in .config. This usage is + broken. As introduced before, external modules should use + kbuild for building and can therefore use the same methods as + in-tree modules when testing for CONFIG_ definitions. -- cgit v0.10.2 From 5793210cb1a431f98c85803920d5ba1440c6246e Mon Sep 17 00:00:00 2001 From: matt mooney Date: Fri, 1 Oct 2010 21:21:55 -0700 Subject: Documentation/kbuild: modules.txt cleanup A few modifications done for consistency, such as adding the shell prompt for command line examples and trailing slash for directories. Also corrects the module include header and fixes a few grammar issues that I introduced. Signed-off-by: matt mooney Signed-off-by: Michal Marek diff --git a/Documentation/kbuild/modules.txt b/Documentation/kbuild/modules.txt index b572db3..3fb39e0 100644 --- a/Documentation/kbuild/modules.txt +++ b/Documentation/kbuild/modules.txt @@ -1,11 +1,11 @@ Building External Modules -This document describes how-to build an out-of-tree kernel module. +This document describes how to build an out-of-tree kernel module. === Table of Contents === 1 Introduction - === 2 How-to Build External Modules + === 2 How to Build External Modules --- 2.1 Command Syntax --- 2.2 Options --- 2.3 Targets @@ -48,9 +48,9 @@ easily accomplished, and a complete example will be presented in section 3. -=== 2. How-to Build External Modules +=== 2. How to Build External Modules -To build external modules, you must have a pre-built kernel available +To build external modules, you must have a prebuilt kernel available that contains the configuration and header files used in the build. Also, the kernel must have been built with modules enabled. If you are using a distribution kernel, there will be a package for the kernel you @@ -69,19 +69,19 @@ executed to make module versioning work. The command to build an external module is: - make -C M=$PWD + $ make -C M=$PWD The kbuild system knows that an external module is being built due to the "M=" option given in the command. To build against the running kernel use: - make -C /lib/modules/`uname -r`/build M=$PWD + $ make -C /lib/modules/`uname -r`/build M=$PWD Then to install the module(s) just built, add the target "modules_install" to the command: - make -C /lib/modules/`uname -r`/build M=$PWD modules_install + $ make -C /lib/modules/`uname -r`/build M=$PWD modules_install --- 2.2 Options @@ -121,7 +121,7 @@ executed to make module versioning work. modules_install Install the external module(s). The default location is - /lib/modules//extra, but a prefix may + /lib/modules//extra/, but a prefix may be added with INSTALL_MOD_PATH (discussed in section 5). clean @@ -164,7 +164,7 @@ needed listing the files: NOTE: Further documentation describing the syntax used by kbuild is located in Documentation/kbuild/makefiles.txt. -The examples below demonstrate how-to create a build file for the +The examples below demonstrate how to create a build file for the module 8123.ko, which is built from the following files: 8123_if.c @@ -205,14 +205,14 @@ module 8123.ko, which is built from the following files: of the makefile. In the example, kbuild will only see the two assignments, whereas "make" will see everything except these two assignments. This is due to two passes made on the file: - the first pass is by the "make" instance run on the - command line; the second pass is by the kbuild system, which is + the first pass is by the "make" instance run on the command + line; the second pass is by the kbuild system, which is initiated by the parameterized "make" in the default target. --- 3.2 Separate Kbuild File and Makefile In newer versions of the kernel, kbuild will first look for a - file named "Kbuild", and only if that is not found, will it + file named "Kbuild," and only if that is not found, will it then look for a makefile. Utilizing a "Kbuild" file allows us to split up the makefile from example 1 into two files: @@ -288,8 +288,8 @@ module 8123.ko, which is built from the following files: --- 3.4 Building Multiple Modules kbuild supports building multiple modules with a single build - file. For example, if you want to build two modules, foo and - bar, the kbuild lines would be: + file. For example, if you wanted to build two modules, foo.ko + and bar.ko, the kbuild lines would be: obj-m := foo.o bar.o foo-y := @@ -320,7 +320,7 @@ according to the following rule: To include a header file located under include/linux/, simply use: - #include + #include kbuild will add options to "gcc" so the relevant directories are searched. @@ -330,7 +330,7 @@ according to the following rule: External modules tend to place header files in a separate include/ directory where their source is located, although this is not the usual kernel style. To inform kbuild of the - directory use either ccflags-y or CFLAGS_.o. + directory, use either ccflags-y or CFLAGS_.o. Using the example from section 3, if we moved 8123_if.h to a subdirectory named include, the resulting kbuild file would @@ -390,11 +390,11 @@ according to the following rule: Modules which are included in the kernel are installed in the directory: - /lib/modules/$(KERNELRELEASE)/kernel + /lib/modules/$(KERNELRELEASE)/kernel/ And external modules are installed in: - /lib/modules/$(KERNELRELEASE)/extra + /lib/modules/$(KERNELRELEASE)/extra/ --- 5.1 INSTALL_MOD_PATH @@ -403,7 +403,7 @@ And external modules are installed in: installation path using the variable INSTALL_MOD_PATH: $ make INSTALL_MOD_PATH=/frodo modules_install - => Install dir: /frodo/lib/modules/$(KERNELRELEASE)/kernel + => Install dir: /frodo/lib/modules/$(KERNELRELEASE)/kernel/ INSTALL_MOD_PATH may be set as an ordinary shell variable or, as shown above, can be specified on the command line when @@ -413,14 +413,14 @@ And external modules are installed in: --- 5.2 INSTALL_MOD_DIR External modules are by default installed to a directory under - /lib/modules/$(KERNELRELEASE)/extra, but you may wish to locate - modules for a specific functionality in a separate directory. - For this purpose, use INSTALL_MOD_DIR to specify an alternative - name to "extra." + /lib/modules/$(KERNELRELEASE)/extra/, but you may wish to + locate modules for a specific functionality in a separate + directory. For this purpose, use INSTALL_MOD_DIR to specify an + alternative name to "extra." $ make INSTALL_MOD_DIR=gandalf -C $KDIR \ M=$PWD modules_install - => Install dir: /lib/modules/$(KERNELRELEASE)/gandalf + => Install dir: /lib/modules/$(KERNELRELEASE)/gandalf/ === 6. Module Versioning @@ -478,9 +478,9 @@ build. Use a top-level kbuild file If you have two modules, foo.ko and bar.ko, where - foo.ko needs symbols from bar.ko, then you can use a + foo.ko needs symbols from bar.ko, you can use a common top-level kbuild file so both modules are - compiled in the same build. Consider following + compiled in the same build. Consider the following directory layout: ./foo/ <= contains foo.ko @@ -491,10 +491,11 @@ build. #./Kbuild (or ./Makefile): obj-y := foo/ bar/ - And executing: + And executing + $ make -C $KDIR M=$PWD - Will then do the expected and compile both modules with + will then do the expected and compile both modules with full knowledge of symbols from either module. Use an extra Module.symvers file @@ -512,10 +513,11 @@ build. Use "make" variable KBUILD_EXTRA_SYMBOLS If it is impractical to copy Module.symvers from another module, you can assign a space separated list - of files to KBUILD_EXTRA_SYMBOLS in your build - file. These files will be loaded by modpost during the + of files to KBUILD_EXTRA_SYMBOLS in your build file. + These files will be loaded by modpost during the initialization of its symbol tables. + === 7. Tips & Tricks --- 7.1 Testing for CONFIG_FOO_BAR -- cgit v0.10.2 From af78f217127dc0b6d364ca67ce128f7aafed99f9 Mon Sep 17 00:00:00 2001 From: Nicolas Palix Date: Fri, 8 Oct 2010 21:27:36 +0200 Subject: MAINTAINERS: Coccinelle: Update email address Update my email address Signed-off-by: Nicolas Palix Signed-off-by: Michal Marek diff --git a/MAINTAINERS b/MAINTAINERS index e4f9e41..146090d 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -1479,7 +1479,7 @@ F: drivers/platform/x86/classmate-laptop.c COCCINELLE/Semantic Patches (SmPL) M: Julia Lawall M: Gilles Muller -M: Nicolas Palix +M: Nicolas Palix L: cocci@diku.dk (moderated for non-subscribers) W: http://coccinelle.lip6.fr/ S: Supported -- cgit v0.10.2 From 32af08987b8f093195f8eef3ee4e1dbcfee891cd Mon Sep 17 00:00:00 2001 From: Nicolas Palix Date: Wed, 13 Oct 2010 00:49:07 +0200 Subject: Coccinelle: Update documentation - Add information about use of the C={1,2} make flag - Add a description of the new chain mode mechanism - Add a link to the wiki Signed-off-by: Nicolas Palix Signed-off-by: Julia Lawall Signed-off-by: Michal Marek diff --git a/Documentation/coccinelle.txt b/Documentation/coccinelle.txt index cd2b028..de51a3e 100644 --- a/Documentation/coccinelle.txt +++ b/Documentation/coccinelle.txt @@ -24,6 +24,9 @@ of many distributions, e.g. : You can get the latest version released from the Coccinelle homepage at http://coccinelle.lip6.fr/ +Information and tips about Coccinelle are also provided on the wiki +pages at http://cocci.ekstranet.diku.dk/wiki/doku.php + Once you have it, run the following command: ./configure @@ -41,20 +44,22 @@ A Coccinelle-specific target is defined in the top level Makefile. This target is named 'coccicheck' and calls the 'coccicheck' front-end in the 'scripts' directory. -Four modes are defined: report, patch, context, and org. The mode to +Four modes are defined: patch, report, context, and org. The mode to use is specified by setting the MODE variable with 'MODE='. +'patch' proposes a fix, when possible. + 'report' generates a list in the following format: file:line:column-column: message -'patch' proposes a fix, when possible. - 'context' highlights lines of interest and their context in a diff-like style.Lines of interest are indicated with '-'. 'org' generates a report in the Org mode format of Emacs. -Note that not all semantic patches implement all modes. +Note that not all semantic patches implement all modes. For easy use +of Coccinelle, the default mode is "chain" which tries the previous +modes in the order above until one succeeds. To make a report for every semantic patch, run the following command: @@ -68,9 +73,9 @@ To produce patches, run: The coccicheck target applies every semantic patch available in the -subdirectories of 'scripts/coccinelle' to the entire Linux kernel. +sub-directories of 'scripts/coccinelle' to the entire Linux kernel. -For each semantic patch, a changelog message is proposed. It gives a +For each semantic patch, a commit message is proposed. It gives a description of the problem being checked by the semantic patch, and includes a reference to Coccinelle. @@ -93,12 +98,35 @@ or make coccicheck COCCI= MODE=report + Using Coccinelle on (modified) files +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + +To apply Coccinelle on a file basis, instead of a directory basis, the +following command may be used: + + make C=1 CHECK="scripts/coccicheck" + +To check only newly edited code, use the value 2 for the C flag, i.e. + + make C=2 CHECK="scripts/coccicheck" + +This runs every semantic patch in scripts/coccinelle by default. The +COCCI variable may additionally be used to only apply a single +semantic patch as shown in the previous section. + +The "chain" mode is the default. You can select another one with the +MODE variable explained above. + +In this mode, there is no information about semantic patches +displayed, and no commit message proposed. + + Proposing new semantic patches ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ New semantic patches can be proposed and submitted by kernel developers. For sake of clarity, they should be organized in the -subdirectories of 'scripts/coccinelle/'. +sub-directories of 'scripts/coccinelle/'. Detailed description of the 'report' mode -- cgit v0.10.2 From 3c908417677f890265ff9d5efd372207cf75fd45 Mon Sep 17 00:00:00 2001 From: Nicolas Palix Date: Fri, 8 Oct 2010 21:27:38 +0200 Subject: Coccinelle: Improve user information with a new kind of comment Improve user information with a new kind of comment about semantic patch output. Fix spelling. Signed-off-by: Nicolas Palix Signed-off-by: Julia Lawall Signed-off-by: Michal Marek diff --git a/scripts/coccicheck b/scripts/coccicheck index b8bcf1f..ef78c87 100755 --- a/scripts/coccicheck +++ b/scripts/coccicheck @@ -25,7 +25,7 @@ fi if [ "$MODE" = "" ] ; then if [ "$ONLINE" = "0" ] ; then - echo 'You have not explicitly specify the mode to use. Fallback to "report".' + echo 'You have not explicitly specified the mode to use. Fallback to "report".' echo 'You can specify the mode with "make coccicheck MODE="' echo 'Available modes are: report, patch, context, org' fi @@ -52,10 +52,12 @@ coccinelle () { FILE=`echo $COCCI | sed "s|$srctree/||"` - echo "Processing `basename $COCCI` with option(s) \"$OPT\"" + echo "Processing `basename $COCCI`" + echo "with option(s) \"$OPT\"" + echo '' echo 'Message example to submit a patch:' - sed -e '/\/\/\//!d' -e 's|^///||' $COCCI + sed -ne 's|^///||p' $COCCI echo ' The semantic patch that makes this change is available' echo " in $FILE." @@ -64,6 +66,12 @@ coccinelle () { echo ' http://coccinelle.lip6.fr/' echo '' + if [ "`sed -ne 's|^//#||p' $COCCI`" ] ; then + echo 'Semantic patch information:' + sed -ne 's|^//#||p' $COCCI + echo '' + fi + $SPATCH -D $MODE $FLAGS -sp_file $COCCI $OPT -dir $srctree || exit 1 else $SPATCH -D $MODE $FLAGS -sp_file $COCCI $OPT $OPTIONS || exit 1 -- cgit v0.10.2 From e90f6590219a0082618b9b6828458b1c74f37cc0 Mon Sep 17 00:00:00 2001 From: Nicolas Palix Date: Fri, 8 Oct 2010 21:27:39 +0200 Subject: Coccinelle: Use new comment format to explain kfree.cocci Use new comment format to separate proposed commit message and information about generated false positives Signed-off-by: Nicolas Palix Signed-off-by: Julia Lawall Signed-off-by: Michal Marek diff --git a/scripts/coccinelle/free/kfree.cocci b/scripts/coccinelle/free/kfree.cocci index c13a539..f9f79d9 100644 --- a/scripts/coccinelle/free/kfree.cocci +++ b/scripts/coccinelle/free/kfree.cocci @@ -1,7 +1,8 @@ -/// Find a use after free. Values of variables may imply that some -/// execution paths are not possible, resulting in false positives. -/// Another source of false positives are macros such as -/// SCTP_DBG_OBJCNT_DEC that do not actually evaluate their argument +/// Find a use after free. +//# Values of variables may imply that some +//# execution paths are not possible, resulting in false positives. +//# Another source of false positives are macros such as +//# SCTP_DBG_OBJCNT_DEC that do not actually evaluate their argument /// // Confidence: Moderate // Copyright: (C) 2010 Nicolas Palix, DIKU. GPLv2. -- cgit v0.10.2 From 2c1160c87465ee2d61eee14596e5376d2ff74f31 Mon Sep 17 00:00:00 2001 From: Nicolas Palix Date: Fri, 8 Oct 2010 21:27:40 +0200 Subject: Coccinelle: Add a new mode named 'chain' spatch now returns -1 when a virtual rule (given with -D on the command line) is not defined in the semantic patch. Using this spatch feature, coccicheck is now tries several modes by default, in the order: patch, report, context, org Signed-off-by: Nicolas Palix Signed-off-by: Julia Lawall Signed-off-by: Michal Marek diff --git a/scripts/coccicheck b/scripts/coccicheck index ef78c87..c7beb63 100755 --- a/scripts/coccicheck +++ b/scripts/coccicheck @@ -16,6 +16,7 @@ if [ "$C" = "1" -o "$C" = "2" ]; then else ONLINE=0 FLAGS="-very_quiet" + OPTIONS="-dir $srctree" fi if [ ! -x "$SPATCH" ]; then @@ -25,11 +26,11 @@ fi if [ "$MODE" = "" ] ; then if [ "$ONLINE" = "0" ] ; then - echo 'You have not explicitly specified the mode to use. Fallback to "report".' + echo 'You have not explicitly specified the mode to use. Using default "chain" mode.' + echo 'All available modes will be tried (in that order): patch, report, context, org' echo 'You can specify the mode with "make coccicheck MODE="' - echo 'Available modes are: report, patch, context, org' fi - MODE="report" + MODE="chain" fi if [ "$ONLINE" = "0" ] ; then @@ -71,10 +72,15 @@ coccinelle () { sed -ne 's|^//#||p' $COCCI echo '' fi + fi - $SPATCH -D $MODE $FLAGS -sp_file $COCCI $OPT -dir $srctree || exit 1 + if [ "$MODE" = "chain" ] ; then + $SPATCH -D patch $FLAGS -sp_file $COCCI $OPT $OPTIONS || \ + $SPATCH -D report $FLAGS -sp_file $COCCI $OPT $OPTIONS || \ + $SPATCH -D context $FLAGS -sp_file $COCCI $OPT $OPTIONS || \ + $SPATCH -D org $FLAGS -sp_file $COCCI $OPT $OPTIONS || exit 1 else - $SPATCH -D $MODE $FLAGS -sp_file $COCCI $OPT $OPTIONS || exit 1 + $SPATCH -D $MODE $FLAGS -sp_file $COCCI $OPT $OPTIONS || exit 1 fi } -- cgit v0.10.2 From 03ee0c42a77fbf60bfa6b0eacaff30ae4e19fce8 Mon Sep 17 00:00:00 2001 From: Nicolas Palix Date: Fri, 8 Oct 2010 21:27:41 +0200 Subject: Coccinelle: Use the -no_show_diff option for org and report mode This allows to write the semantic patches with code sharing for the matching parts. Signed-off-by: Nicolas Palix Signed-off-by: Julia Lawall Signed-off-by: Michal Marek diff --git a/scripts/coccicheck b/scripts/coccicheck index c7beb63..7529af1 100755 --- a/scripts/coccicheck +++ b/scripts/coccicheck @@ -31,6 +31,8 @@ if [ "$MODE" = "" ] ; then echo 'You can specify the mode with "make coccicheck MODE="' fi MODE="chain" +elif [ "$MODE" = "report" -o "$MODE" = "org" ] ; then + FLAGS="$FLAGS -no_show_diff" fi if [ "$ONLINE" = "0" ] ; then @@ -75,10 +77,10 @@ coccinelle () { fi if [ "$MODE" = "chain" ] ; then - $SPATCH -D patch $FLAGS -sp_file $COCCI $OPT $OPTIONS || \ - $SPATCH -D report $FLAGS -sp_file $COCCI $OPT $OPTIONS || \ - $SPATCH -D context $FLAGS -sp_file $COCCI $OPT $OPTIONS || \ - $SPATCH -D org $FLAGS -sp_file $COCCI $OPT $OPTIONS || exit 1 + $SPATCH -D patch $FLAGS -sp_file $COCCI $OPT $OPTIONS || \ + $SPATCH -D report $FLAGS -sp_file $COCCI $OPT $OPTIONS -no_show_diff || \ + $SPATCH -D context $FLAGS -sp_file $COCCI $OPT $OPTIONS || \ + $SPATCH -D org $FLAGS -sp_file $COCCI $OPT $OPTIONS -no_show_diff || exit 1 else $SPATCH -D $MODE $FLAGS -sp_file $COCCI $OPT $OPTIONS || exit 1 fi -- cgit v0.10.2 From c25f415751c0c5507561d997fe5f7f05f4342912 Mon Sep 17 00:00:00 2001 From: Amerigo Wang Date: Wed, 27 Oct 2010 12:42:00 -0700 Subject: scripts/namespace.pl: fix wrong source path File::Find will do chdir automatically, so we need to get the absolute patch with $File::Find::dir. Reported-by: Stephen Hemminger Signed-off-by: Amerigo Wang Cc: Michal Marek Signed-off-by: Andrew Morton Signed-off-by: Michal Marek diff --git a/scripts/namespace.pl b/scripts/namespace.pl index 361d0f7..fb4e245 100755 --- a/scripts/namespace.pl +++ b/scripts/namespace.pl @@ -167,11 +167,11 @@ sub do_nm printf STDERR "$fullname is not an object file\n"; return; } - ($source = $fullname) =~ s/\.o$//; - if (-e "$objtree$source.c" || -e "$objtree$source.S") { - $source = "$objtree$source"; + ($source = $basename) =~ s/\.o$//; + if (-e "$source.c" || -e "$source.S") { + $source = "$objtree$File::Find::dir/$source"; } else { - $source = "$srctree$source"; + $source = "$srctree$File::Find::dir/$source"; } if (! -e "$source.c" && ! -e "$source.S") { # No obvious source, exclude the object if it is conglomerate -- cgit v0.10.2 From 43f683c9e465a64259c6058a7c313facc697b203 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Wed, 27 Oct 2010 12:42:00 -0700 Subject: scripts/namespace.pl: update file exclusion list Signed-off-by: Stephen Hemminger Signed-off-by: Amerigo Wang Cc: Michal Marek Signed-off-by: Andrew Morton Signed-off-by: Michal Marek diff --git a/scripts/namespace.pl b/scripts/namespace.pl index fb4e245..c8d0922 100755 --- a/scripts/namespace.pl +++ b/scripts/namespace.pl @@ -84,6 +84,58 @@ my %ksymtab = (); # names that appear in __ksymtab_ my %ref = (); # $ref{$name} exists if there is a true external reference to $name my %export = (); # $export{$name} exists if there is an EXPORT_... of $name +my %nmexception = ( + 'fs/ext3/bitmap' => 1, + 'fs/ext4/bitmap' => 1, + 'arch/x86/lib/thunk_32' => 1, + 'arch/x86/lib/cmpxchg' => 1, + 'arch/x86/vdso/vdso32/note' => 1, + 'lib/irq_regs' => 1, + 'usr/initramfs_data' => 1, + 'drivers/scsi/aic94xx/aic94xx_dump' => 1, + 'drivers/scsi/libsas/sas_dump' => 1, + 'lib/dec_and_lock' => 1, + 'drivers/ide/ide-probe-mini' => 1, + 'usr/initramfs_data' => 1, + 'drivers/acpi/acpia/exdump' => 1, + 'drivers/acpi/acpia/rsdump' => 1, + 'drivers/acpi/acpia/nsdumpdv' => 1, + 'drivers/acpi/acpia/nsdump' => 1, + 'arch/ia64/sn/kernel/sn2/io' => 1, + 'arch/ia64/kernel/gate-data' => 1, + 'security/capability' => 1, + 'fs/ntfs/sysctl' => 1, + 'fs/jfs/jfs_debug' => 1, +); + +my %nameexception = ( + 'mod_use_count_' => 1, + '__initramfs_end' => 1, + '__initramfs_start' => 1, + '_einittext' => 1, + '_sinittext' => 1, + 'kallsyms_names' => 1, + 'kallsyms_num_syms' => 1, + 'kallsyms_addresses'=> 1, + '__this_module' => 1, + '_etext' => 1, + '_edata' => 1, + '_end' => 1, + '__bss_start' => 1, + '_text' => 1, + '_stext' => 1, + '__gp' => 1, + 'ia64_unw_start' => 1, + 'ia64_unw_end' => 1, + '__init_begin' => 1, + '__init_end' => 1, + '__bss_stop' => 1, + '__nosave_begin' => 1, + '__nosave_end' => 1, + 'pg0' => 1, +); + + &find(\&linux_objects, '.'); # find the objects and do_nm on them &list_multiply_defined(); &resolve_external_references(); @@ -270,27 +322,9 @@ sub do_nm close($nmdata); if ($#nmdata < 0) { - if ( - $fullname ne "lib/brlock.o" - && $fullname ne "lib/dec_and_lock.o" - && $fullname ne "fs/xfs/xfs_macros.o" - && $fullname ne "drivers/ide/ide-probe-mini.o" - && $fullname ne "usr/initramfs_data.o" - && $fullname ne "drivers/acpi/executer/exdump.o" - && $fullname ne "drivers/acpi/resources/rsdump.o" - && $fullname ne "drivers/acpi/namespace/nsdumpdv.o" - && $fullname ne "drivers/acpi/namespace/nsdump.o" - && $fullname ne "arch/ia64/sn/kernel/sn2/io.o" - && $fullname ne "arch/ia64/kernel/gate-data.o" - && $fullname ne "drivers/ieee1394/oui.o" - && $fullname ne "security/capability.o" - && $fullname ne "sound/core/wrappers.o" - && $fullname ne "fs/ntfs/sysctl.o" - && $fullname ne "fs/jfs/jfs_debug.o" - ) { - printf "No nm data for $fullname\n"; - } - return; + printf "No nm data for $fullname\n" + unless $nmexception{$fullname}; + return; } $nmdata{$fullname} = \@nmdata; } @@ -372,31 +406,7 @@ sub resolve_external_references $ref{$name} = "" } } - elsif ( $name ne "mod_use_count_" - && $name ne "__initramfs_end" - && $name ne "__initramfs_start" - && $name ne "_einittext" - && $name ne "_sinittext" - && $name ne "kallsyms_names" - && $name ne "kallsyms_num_syms" - && $name ne "kallsyms_addresses" - && $name ne "__this_module" - && $name ne "_etext" - && $name ne "_edata" - && $name ne "_end" - && $name ne "__bss_start" - && $name ne "_text" - && $name ne "_stext" - && $name ne "__gp" - && $name ne "ia64_unw_start" - && $name ne "ia64_unw_end" - && $name ne "__init_begin" - && $name ne "__init_end" - && $name ne "__bss_stop" - && $name ne "__nosave_begin" - && $name ne "__nosave_end" - && $name ne "pg0" - && $name ne "__module_text_address" + elsif ( ! $nameexception{$name} && $name !~ /^__sched_text_/ && $name !~ /^__start_/ && $name !~ /^__end_/ @@ -407,7 +417,6 @@ sub resolve_external_references && $name !~ /^__.*per_cpu_end/ && $name !~ /^__alt_instructions/ && $name !~ /^__setup_/ - && $name !~ /^jiffies/ && $name !~ /^__mod_timer/ && $name !~ /^__mod_page_state/ && $name !~ /^init_module/ -- cgit v0.10.2 From e8cf981346b78ee50c2bfce83be9ee55704b3d4f Mon Sep 17 00:00:00 2001 From: Amerigo Wang Date: Wed, 27 Oct 2010 12:42:01 -0700 Subject: scripts/namespace.pl: some bug fixes 1. Teach namespace.pl to understand "V" and "v" 2. cond_syscalls are moved into kernel/sys_ni.c Signed-off-by: Amerigo Wang Cc: Stephen Hemminger Cc: Michal Marek Signed-off-by: Andrew Morton Signed-off-by: Michal Marek diff --git a/scripts/namespace.pl b/scripts/namespace.pl index c8d0922..bccf610 100755 --- a/scripts/namespace.pl +++ b/scripts/namespace.pl @@ -266,6 +266,7 @@ sub do_nm # T global label/procedure # U external reference # W weak external reference to text that has been resolved + # V similar to W, but the value of the weak symbol becomes zero with no error. # a assembler equate # b static variable, uninitialised # d static variable, initialised @@ -274,8 +275,9 @@ sub do_nm # s static variable, uninitialised, small bss # t static label/procedures # w weak external reference to text that has not been resolved + # v similar to w # ? undefined type, used a lot by modules - if ($type !~ /^[ABCDGRSTUWabdgrstw?]$/) { + if ($type !~ /^[ABCDGRSTUWVabdgrstwv?]$/) { printf STDERR "nm output for $fullname contains unknown type '$_'\n"; } elsif ($name =~ /\./) { @@ -286,7 +288,7 @@ sub do_nm # binutils keeps changing the type for exported symbols, force it to R $type = 'R' if ($name =~ /^__ksymtab/ || $name =~ /^__kstrtab/); $name =~ s/_R[a-f0-9]{8}$//; # module versions adds this - if ($type =~ /[ABCDGRSTW]/ && + if ($type =~ /[ABCDGRSTWV]/ && $name ne 'init_module' && $name ne 'cleanup_module' && $name ne 'Using_Versions' && @@ -353,11 +355,12 @@ sub list_multiply_defined foreach my $name (keys(%def)) { if ($#{$def{$name}} > 0) { # Special case for cond_syscall - if ($#{$def{$name}} == 1 && $name =~ /^sys_/ && - ($def{$name}[0] eq "kernel/sys.o" || - $def{$name}[1] eq "kernel/sys.o")) { - &drop_def("kernel/sys.o", $name); - next; + if ($#{$def{$name}} == 1 && $name =~ /^sys_/) { + if($def{$name}[0] eq "kernel/sys_ni.o" || + $def{$name}[1] eq "kernel/sys_ni.o") { + &drop_def("kernel/sys_ni.o", $name); + next; + } } # Special case for i386 entry code if ($#{$def{$name}} == 1 && $name =~ /^__kernel_/ && -- cgit v0.10.2 From abb438526201c6a79949ad45375c051b6681c253 Mon Sep 17 00:00:00 2001 From: Amerigo Wang Date: Wed, 27 Oct 2010 12:42:01 -0700 Subject: scripts/namespace.pl: improve to get more correct results Exclude more symbols from arch/x86/vdso/ and arch/x86/boot/; add some more linker-defined symbols into exception list; add other cond_syscalls besides "sys_*". Signed-off-by: Amerigo Wang Cc: Stephen Hemminger Cc: Michal Marek Signed-off-by: Andrew Morton Signed-off-by: Michal Marek diff --git a/scripts/namespace.pl b/scripts/namespace.pl index bccf610..a71be6b 100755 --- a/scripts/namespace.pl +++ b/scripts/namespace.pl @@ -133,6 +133,12 @@ my %nameexception = ( '__nosave_begin' => 1, '__nosave_end' => 1, 'pg0' => 1, + 'vdso_enabled' => 1, + '__stack_chk_fail' => 1, + 'VDSO32_PRELINK' => 1, + 'VDSO32_vsyscall' => 1, + 'VDSO32_rt_sigreturn'=>1, + 'VDSO32_sigreturn' => 1, ); @@ -157,7 +163,8 @@ sub linux_objects if (/.*\.o$/ && ! ( m:/built-in.o$: - || m:arch/x86/kernel/vsyscall-syms.o$: + || m:arch/x86/vdso/: + || m:arch/x86/boot/: || m:arch/ia64/ia32/ia32.o$: || m:arch/ia64/kernel/gate-syms.o$: || m:arch/ia64/lib/__divdi3.o$: @@ -200,6 +207,7 @@ sub linux_objects || m:^.*/\.tmp_: || m:^\.tmp_: || m:/vmlinux-obj.o$: + || m:^tools/: ) ) { do_nm($basename, $_); @@ -355,20 +363,15 @@ sub list_multiply_defined foreach my $name (keys(%def)) { if ($#{$def{$name}} > 0) { # Special case for cond_syscall - if ($#{$def{$name}} == 1 && $name =~ /^sys_/) { + if ($#{$def{$name}} == 1 && + ($name =~ /^sys_/ || $name =~ /^compat_sys_/ || + $name =~ /^sys32_/)) { if($def{$name}[0] eq "kernel/sys_ni.o" || $def{$name}[1] eq "kernel/sys_ni.o") { &drop_def("kernel/sys_ni.o", $name); next; } } - # Special case for i386 entry code - if ($#{$def{$name}} == 1 && $name =~ /^__kernel_/ && - $def{$name}[0] eq "arch/x86/kernel/vsyscall-int80_32.o" && - $def{$name}[1] eq "arch/x86/kernel/vsyscall-sysenter_32.o") { - &drop_def("arch/x86/kernel/vsyscall-sysenter_32.o", $name); - next; - } printf "$name is multiply defined in :-\n"; foreach my $module (@{$def{$name}}) { -- cgit v0.10.2 From d0f95c782686dbfbb415d533881a06c2bde17ee0 Mon Sep 17 00:00:00 2001 From: Namhyung Kim Date: Fri, 22 Oct 2010 23:32:10 +0900 Subject: kbuild: check return value of asprintf() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Check return value of asprintf() in docsect() and exit if error occurs. This removes following warning: HOSTCC scripts/basic/docproc scripts/basic/docproc.c: In function ‘docsect’: scripts/basic/docproc.c:336: warning: ignoring return value of ‘asprintf’, declared with attribute warn_unused_result Signed-off-by: Namhyung Kim Acked-by: Randy Dunlap Signed-off-by: Michal Marek diff --git a/scripts/basic/docproc.c b/scripts/basic/docproc.c index fc3b18d..98dec87 100644 --- a/scripts/basic/docproc.c +++ b/scripts/basic/docproc.c @@ -333,7 +333,10 @@ static void docsect(char *filename, char *line) if (*s == '\n') *s = '\0'; - asprintf(&s, "DOC: %s", line); + if (asprintf(&s, "DOC: %s", line) < 0) { + perror("asprintf"); + exit(1); + } consume_symbol(s); free(s); -- cgit v0.10.2 From 532cf2907ac3b9c2345d76251764f4f4e602c921 Mon Sep 17 00:00:00 2001 From: Dick Streefland Date: Sat, 23 Oct 2010 00:02:44 +0200 Subject: scripts/extract-ikconfig: add support for bzip2, lzma and lzo Add support for kernels compressed with bzip2, lzma or lzo to the extract-ikconfig script. Fixes kernel bugzilla #19852: https://bugzilla.kernel.org/show_bug.cgi?id=19852 Signed-off-by: Dick Streefland Tested-by: Justin Signed-off-by: Michal Marek diff --git a/scripts/extract-ikconfig b/scripts/extract-ikconfig index 37f30d3..1512c0a 100755 --- a/scripts/extract-ikconfig +++ b/scripts/extract-ikconfig @@ -7,12 +7,10 @@ # The obscure use of the "tr" filter is to work around older versions of # "grep" that report the byte offset of the line instead of the pattern. # -# (c) 2009, Dick Streefland +# (c) 2009,2010 Dick Streefland # Licensed under the terms of the GNU General Public License. # ---------------------------------------------------------------------- -gz1='\037\213\010' -gz2='01' cf1='IKCFG_ST\037\213\010' cf2='0123456789' @@ -21,11 +19,25 @@ dump_config() if pos=`tr "$cf1\n$cf2" "\n$cf2=" < "$1" | grep -abo "^$cf2"` then pos=${pos%%:*} - tail -c+$(($pos+8)) "$1" | zcat -q - exit 0 + tail -c+$(($pos+8)) "$1" | zcat > $tmp1 2> /dev/null + if [ $? != 1 ] + then # exit status must be 0 or 2 (trailing garbage warning) + cat $tmp1 + exit 0 + fi fi } +try_decompress() +{ + for pos in `tr "$1\n$2" "\n$2=" < "$img" | grep -abo "^$2"` + do + pos=${pos%%:*} + tail -c+$pos "$img" | $3 > $tmp2 2> /dev/null + dump_config $tmp2 + done +} + # Check invocation: me=${0##*/} img=$1 @@ -35,18 +47,19 @@ then exit 2 fi +# Prepare temp files: +tmp1=/tmp/ikconfig$$.1 +tmp2=/tmp/ikconfig$$.2 +trap "rm -f $tmp1 $tmp2" 0 + # Initial attempt for uncompressed images or objects: dump_config "$img" -# That didn't work, so decompress and try again: -tmp=/tmp/ikconfig$$ -trap "rm -f $tmp" 0 -for pos in `tr "$gz1\n$gz2" "\n$gz2=" < "$img" | grep -abo "^$gz2"` -do - pos=${pos%%:*} - tail -c+$pos "$img" | zcat 2> /dev/null > $tmp - dump_config $tmp -done +# That didn't work, so retry after decompression. +try_decompress '\037\213\010' xy gunzip +try_decompress 'BZh' xy bunzip2 +try_decompress '\135\0\0\0' xxx unlzma +try_decompress '\211\114\132' xy 'lzop -d' # Bail out: echo "$me: Cannot find kernel config." >&2 -- cgit v0.10.2 From 24a54f7974a616385b96cd939e004592e2cea484 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Wed, 13 Oct 2010 15:58:30 -0700 Subject: namespace: add source file location exceptions Teach namespace checker about some special case files where the source is in unusual location. This fixes many of the source file not found errors (more can be added), and also prevents false positives for functions not being used. Signed-off-by: Stephen Hemminger Signed-off-by: Michal Marek diff --git a/scripts/namespace.pl b/scripts/namespace.pl index a71be6b..fd2d946 100755 --- a/scripts/namespace.pl +++ b/scripts/namespace.pl @@ -141,6 +141,26 @@ my %nameexception = ( 'VDSO32_sigreturn' => 1, ); +# Files with exceptions to source file location +my %sourceloc = ( + 'net/dccp/dccp_probe.o' => 'probe', + 'net/dccp/dccp_ipv4.o' => 'ipv4', + 'net/dccp/dccp_ipv6.o' => 'ipv6', + 'net/dccp/dccp_diag.o' => 'diag', + 'drivers/char/hw_random/rng-core.o' => 'core', + 'fs/fat/msdos.o' => 'namei_msdos', + 'fs/fat/vfat.o' => 'namei_vfat', + 'fs/nfs_common/nfs_acl.o' => 'nfsacl', + 'sound/soundcore.o' => 'sound_core', + 'drivers/md/dm-mirror.o' => 'dm-raid1', + 'drivers/message/i2o/i2o_bus.o' => 'bus-osm', + 'arch/x86/kvm/kvm-amd.o' => 'svm', + 'arch/x86/kvm/kvm-intel.o' => 'vmx', + 'arch/x86/crypto/twofish-x86_64.o' => 'twofish-x86_64-asm_64', + 'arch/x86/crypto/aes-x86_64.o' => 'aes-x86_64-asm_64', + 'arch/x86/crypto/aesni-intel.o' => 'aesni-intel_asm', + 'arch/x86/crypto/salsa20-x86_64.o' => 'salsa20-x86_64-asm_64', +); &find(\&linux_objects, '.'); # find the objects and do_nm on them &list_multiply_defined(); @@ -228,11 +248,15 @@ sub do_nm return; } ($source = $basename) =~ s/\.o$//; + + $source = $sourceloc{$fullname} if ($sourceloc{$fullname}); + if (-e "$source.c" || -e "$source.S") { $source = "$objtree$File::Find::dir/$source"; } else { $source = "$srctree$File::Find::dir/$source"; } + if (! -e "$source.c" && ! -e "$source.S") { # No obvious source, exclude the object if it is conglomerate open(my $objdumpdata, "$objdump $basename|") -- cgit v0.10.2 From 10247179a7af4f000468e12ae73f3cf657b66235 Mon Sep 17 00:00:00 2001 From: Julia Lawall Date: Sun, 24 Oct 2010 23:37:31 +0200 Subject: Coccinelle: Find nested lock+irqsave functions that use the same flags variables. Signed-off-by: Julia Lawall Signed-off-by: Nicolas Palix Signed-off-by: Michal Marek diff --git a/scripts/coccinelle/locks/flags.cocci b/scripts/coccinelle/locks/flags.cocci new file mode 100644 index 0000000..b4344d8 --- /dev/null +++ b/scripts/coccinelle/locks/flags.cocci @@ -0,0 +1,80 @@ +/// Find nested lock+irqsave functions that use the same flags variables +/// +// Confidence: High +// Copyright: (C) 2010 Nicolas Palix, DIKU. GPLv2. +// Copyright: (C) 2010 Julia Lawall, DIKU. GPLv2. +// Copyright: (C) 2010 Gilles Muller, INRIA/LiP6. GPLv2. +// URL: http://coccinelle.lip6.fr/ +// Comments: +// Options: -no_includes -include_headers + +virtual context +virtual org +virtual report + +@r@ +expression lock1,lock2,flags; +position p1,p2; +@@ + +( +spin_lock_irqsave@p1(lock1,flags) +| +read_lock_irqsave@p1(lock1,flags) +| +write_lock_irqsave@p1(lock1,flags) +) +... when != flags +( +spin_lock_irqsave(lock1,flags) +| +read_lock_irqsave(lock1,flags) +| +write_lock_irqsave(lock1,flags) +| +spin_lock_irqsave@p2(lock2,flags) +| +read_lock_irqsave@p2(lock2,flags) +| +write_lock_irqsave@p2(lock2,flags) +) + +@d@ +expression f <= r.flags; +expression lock1,lock2,flags; +position r.p1, r.p2; +@@ + +( +*spin_lock_irqsave@p1(lock1,flags) +| +*read_lock_irqsave@p1(lock1,flags) +| +*write_lock_irqsave@p1(lock1,flags) +) +... when != f +( +*spin_lock_irqsave@p2(lock2,flags) +| +*read_lock_irqsave@p2(lock2,flags) +| +*write_lock_irqsave@p2(lock2,flags) +) + +// ---------------------------------------------------------------------- + +@script:python depends on d && org@ +p1 << r.p1; +p2 << r.p2; +@@ + +cocci.print_main("original lock",p1) +cocci.print_secs("nested lock+irqsave that reuses flags",p2) + +@script:python depends on d && report@ +p1 << r.p1; +p2 << r.p2; +@@ + +msg="ERROR: nested lock+irqsave that reuses flags from %s." % (p1[0].line) +coccilib.report.print_report(p2[0], msg) -- cgit v0.10.2 From 97c1cf8fdbce332bfacc3e38aae3fe2af1369906 Mon Sep 17 00:00:00 2001 From: Julia Lawall Date: Sun, 24 Oct 2010 23:37:32 +0200 Subject: Coccinelle: Find doubled arguments to boolean or bit operators. Signed-off-by: Julia Lawall Signed-off-by: Nicolas Palix Signed-off-by: Michal Marek diff --git a/scripts/coccinelle/tests/doublebitand.cocci b/scripts/coccinelle/tests/doublebitand.cocci new file mode 100644 index 0000000..9ba73d0 --- /dev/null +++ b/scripts/coccinelle/tests/doublebitand.cocci @@ -0,0 +1,54 @@ +/// Find bit operations that include the same argument more than once +//# One source of false positives is when the argument performs a side +//# effect. Another source of false positives is when a neutral value +//# such as 0 for | is used to indicate no information, to maintain the +//# same structure as other similar expressions +/// +// Confidence: Moderate +// Copyright: (C) 2010 Nicolas Palix, DIKU. GPLv2. +// Copyright: (C) 2010 Julia Lawall, DIKU. GPLv2. +// Copyright: (C) 2010 Gilles Muller, INRIA/LiP6. GPLv2. +// URL: http://coccinelle.lip6.fr/ +// Comments: +// Options: -no_includes -include_headers + +virtual context +virtual org +virtual report + +@r expression@ +expression E; +position p; +@@ + +( +* E@p + & ... & E +| +* E@p + | ... | E +| +* E@p + & ... & !E +| +* E@p + | ... | !E +| +* !E@p + & ... & E +| +* !E@p + | ... | E +) + +@script:python depends on org@ +p << r.p; +@@ + +cocci.print_main("duplicated argument to & or |",p) + +@script:python depends on report@ +p << r.p; +@@ + +coccilib.report.print_report(p[0],"duplicated argument to & or |") diff --git a/scripts/coccinelle/tests/doubletest.cocci b/scripts/coccinelle/tests/doubletest.cocci new file mode 100644 index 0000000..13a2c0e --- /dev/null +++ b/scripts/coccinelle/tests/doubletest.cocci @@ -0,0 +1,40 @@ +/// Find &&/|| operations that include the same argument more than once +//# A common source of false positives is when the argument performs a side +//# effect. +/// +// Confidence: Moderate +// Copyright: (C) 2010 Nicolas Palix, DIKU. GPLv2. +// Copyright: (C) 2010 Julia Lawall, DIKU. GPLv2. +// Copyright: (C) 2010 Gilles Muller, INRIA/LiP6. GPLv2. +// URL: http://coccinelle.lip6.fr/ +// Comments: +// Options: -no_includes -include_headers + +virtual context +virtual org +virtual report + +@r expression@ +expression E; +position p; +@@ + +( +* E@p + || ... || E +| +* E@p + && ... && E +) + +@script:python depends on org@ +p << r.p; +@@ + +cocci.print_main("duplicated argument to && or ||",p) + +@script:python depends on report@ +p << r.p; +@@ + +coccilib.report.print_report(p[0],"duplicated argument to && or ||") -- cgit v0.10.2 From 9dcf7990c2b8afe865d59f809a35a84d10f241f8 Mon Sep 17 00:00:00 2001 From: Nicolas Palix Date: Sun, 24 Oct 2010 23:37:33 +0200 Subject: Coccinelle: Fix documentation A file used as example has been moved elsewhere. Update the documentation accordingly Signed-off-by: Nicolas Palix Reported-by: Julia Lawall Signed-off-by: Michal Marek diff --git a/Documentation/coccinelle.txt b/Documentation/coccinelle.txt index de51a3e..4a276ea 100644 --- a/Documentation/coccinelle.txt +++ b/Documentation/coccinelle.txt @@ -139,7 +139,7 @@ Example: Running - make coccicheck MODE=report COCCI=scripts/coccinelle/err_cast.cocci + make coccicheck MODE=report COCCI=scripts/coccinelle/api/err_cast.cocci will execute the following part of the SmPL script. @@ -177,7 +177,7 @@ identified. Example: Running - make coccicheck MODE=patch COCCI=scripts/coccinelle/err_cast.cocci + make coccicheck MODE=patch COCCI=scripts/coccinelle/api/err_cast.cocci will execute the following part of the SmPL script. @@ -221,7 +221,7 @@ NOTE: The diff-like output generated is NOT an applicable patch. The Example: Running - make coccicheck MODE=context COCCI=scripts/coccinelle/err_cast.cocci + make coccicheck MODE=context COCCI=scripts/coccinelle/api/err_cast.cocci will execute the following part of the SmPL script. @@ -256,7 +256,7 @@ diff -u -p /home/user/linux/crypto/ctr.c /tmp/nothing Example: Running - make coccicheck MODE=org COCCI=scripts/coccinelle/err_cast.cocci + make coccicheck MODE=org COCCI=scripts/coccinelle/api/err_cast.cocci will execute the following part of the SmPL script. -- cgit v0.10.2 From 062c1825a5f6673746efe5f0a5ea3803b0c316a9 Mon Sep 17 00:00:00 2001 From: Nicolas Palix Date: Sun, 24 Oct 2010 23:37:34 +0200 Subject: Coccinelle: Add contextual message Change the message displayed to the user according to the current mode used. Fix trailing white-space and spelling Signed-off-by: Nicolas Palix Signed-off-by: Michal Marek diff --git a/scripts/coccicheck b/scripts/coccicheck index 7529af1..1bb1a1b 100755 --- a/scripts/coccicheck +++ b/scripts/coccicheck @@ -32,7 +32,7 @@ if [ "$MODE" = "" ] ; then fi MODE="chain" elif [ "$MODE" = "report" -o "$MODE" = "org" ] ; then - FLAGS="$FLAGS -no_show_diff" + FLAGS="$FLAGS -no_show_diff" fi if [ "$ONLINE" = "0" ] ; then @@ -47,7 +47,7 @@ coccinelle () { OPT=`grep "Option" $COCCI | cut -d':' -f2` -# The option '-parse_cocci' can be used to syntaxically check the SmPL files. +# The option '-parse_cocci' can be used to syntactically check the SmPL files. # # $SPATCH -D $MODE $FLAGS -parse_cocci $COCCI $OPT > /dev/null @@ -62,7 +62,17 @@ coccinelle () { sed -ne 's|^///||p' $COCCI - echo ' The semantic patch that makes this change is available' + if [ "$MODE" = "patch" ] ; then + echo ' The semantic patch that makes this change is available' + elif [ "$MODE" = "report" ] ; then + echo ' The semantic patch that makes this report is available' + elif [ "$MODE" = "context" ] ; then + echo ' The semantic patch that spots this code is available' + elif [ "$MODE" = "org" ] ; then + echo ' The semantic patch that makes this Org report is available' + else + echo ' The semantic patch that makes this output is available' + fi echo " in $FILE." echo '' echo ' More information about semantic patching is available at' -- cgit v0.10.2 From 9231d9e02a1f92b52bbb1e4474bfd1903835a993 Mon Sep 17 00:00:00 2001 From: Michal Marek Date: Thu, 28 Oct 2010 00:59:56 +0200 Subject: Revert "namespace: add source file location exceptions" This reverts commit 24a54f7974a616385b96cd939e004592e2cea484. Stephen Hemminger writes: > That patch should not be included. It causes more problems than it > solves, since then there are duplicate file locations which causes > false duplicate symbol reports. Reported-by: Stephen Hemminger Signed-off-by: Michal Marek diff --git a/scripts/namespace.pl b/scripts/namespace.pl index fd2d946..a71be6b 100755 --- a/scripts/namespace.pl +++ b/scripts/namespace.pl @@ -141,26 +141,6 @@ my %nameexception = ( 'VDSO32_sigreturn' => 1, ); -# Files with exceptions to source file location -my %sourceloc = ( - 'net/dccp/dccp_probe.o' => 'probe', - 'net/dccp/dccp_ipv4.o' => 'ipv4', - 'net/dccp/dccp_ipv6.o' => 'ipv6', - 'net/dccp/dccp_diag.o' => 'diag', - 'drivers/char/hw_random/rng-core.o' => 'core', - 'fs/fat/msdos.o' => 'namei_msdos', - 'fs/fat/vfat.o' => 'namei_vfat', - 'fs/nfs_common/nfs_acl.o' => 'nfsacl', - 'sound/soundcore.o' => 'sound_core', - 'drivers/md/dm-mirror.o' => 'dm-raid1', - 'drivers/message/i2o/i2o_bus.o' => 'bus-osm', - 'arch/x86/kvm/kvm-amd.o' => 'svm', - 'arch/x86/kvm/kvm-intel.o' => 'vmx', - 'arch/x86/crypto/twofish-x86_64.o' => 'twofish-x86_64-asm_64', - 'arch/x86/crypto/aes-x86_64.o' => 'aes-x86_64-asm_64', - 'arch/x86/crypto/aesni-intel.o' => 'aesni-intel_asm', - 'arch/x86/crypto/salsa20-x86_64.o' => 'salsa20-x86_64-asm_64', -); &find(\&linux_objects, '.'); # find the objects and do_nm on them &list_multiply_defined(); @@ -248,15 +228,11 @@ sub do_nm return; } ($source = $basename) =~ s/\.o$//; - - $source = $sourceloc{$fullname} if ($sourceloc{$fullname}); - if (-e "$source.c" || -e "$source.S") { $source = "$objtree$File::Find::dir/$source"; } else { $source = "$srctree$File::Find::dir/$source"; } - if (! -e "$source.c" && ! -e "$source.S") { # No obvious source, exclude the object if it is conglomerate open(my $objdumpdata, "$objdump $basename|") -- cgit v0.10.2