Date: Wed, 6 Sep 2023 14:01:33 GMT From: Mark Johnston <markj@FreeBSD.org> To: src-committers@FreeBSD.org, dev-commits-src-all@FreeBSD.org, dev-commits-src-main@FreeBSD.org Subject: git: 8527bb2aee6c - main - dtrace: Fix a kernel panic in printm() Message-ID: <202309061401.386E1X2p030800@gitrepo.freebsd.org>
next in thread | raw e-mail | index | archive | help
The branch main has been updated by markj: URL: https://cgit.FreeBSD.org/src/commit/?id=8527bb2aee6ca9013c34445de88217a954b6628d commit 8527bb2aee6ca9013c34445de88217a954b6628d Author: Domagoj Stolfa <domagoj.stolfa@gmail.com> AuthorDate: 2023-09-06 13:25:00 +0000 Commit: Mark Johnston <markj@FreeBSD.org> CommitDate: 2023-09-06 14:00:59 +0000 dtrace: Fix a kernel panic in printm() When using printm(), one should always pass a scratch pointer to it. This is achieved by calling printm with memref BEGIN { printm(fixed_len, memref(ptr, var_len)); } which will return a pointer to the DTrace scratch space of size sizeof(uintptr_t) * 2. However, one can easily call printm() as follows BEGIN { printm(10, (void *)NULL); } and panic the kernel as a result. This commit does two things: (1) adds a new macro DTRACE_INSCRATCHPTR(mstate, ptr, howmany) which checks if a certain pointer is in the DTrace scratch space; (2) uses DTRACE_INSCRATCHPTR() to implement a check on printm()'s DIFO return value in order to avoid the panic and sets CPU_DTRACE_BADADDR if the address is not in the scratch space. Reviewed by: markj MFC after: 2 weeks Differential Revision: https://reviews.freebsd.org/D41722 --- .../contrib/opensolaris/uts/common/dtrace/dtrace.c | 22 +++++++++++++++++++++- 1 file changed, 21 insertions(+), 1 deletion(-) diff --git a/sys/cddl/contrib/opensolaris/uts/common/dtrace/dtrace.c b/sys/cddl/contrib/opensolaris/uts/common/dtrace/dtrace.c index 02a95bfab726..ce02676e0dc1 100644 --- a/sys/cddl/contrib/opensolaris/uts/common/dtrace/dtrace.c +++ b/sys/cddl/contrib/opensolaris/uts/common/dtrace/dtrace.c @@ -515,6 +515,11 @@ do { \ ((mstate)->dtms_scratch_base + (mstate)->dtms_scratch_size - \ (mstate)->dtms_scratch_ptr >= (alloc_sz)) +#define DTRACE_INSCRATCHPTR(mstate, ptr, howmany) \ + ((ptr) >= (mstate)->dtms_scratch_base && \ + (ptr) <= \ + ((mstate)->dtms_scratch_base + (mstate)->dtms_scratch_size - (howmany))) + #define DTRACE_LOADFUNC(bits) \ /*CSTYLED*/ \ uint##bits##_t \ @@ -7739,9 +7744,24 @@ dtrace_probe(dtrace_id_t id, uintptr_t arg0, uintptr_t arg1, } case DTRACEACT_PRINTM: { - /* The DIF returns a 'memref'. */ + /* + * printm() assumes that the DIF returns a + * pointer returned by memref(). memref() is a + * subroutine that is used to get around the + * single-valued returns of DIF and is assumed + * to always be allocated in the scratch space. + * Therefore, we need to validate that the + * pointer given to printm() is in the scratch + * space in order to avoid a potential panic. + */ uintptr_t *memref = (uintptr_t *)(uintptr_t) val; + if (!DTRACE_INSCRATCHPTR(&mstate, + (uintptr_t)memref, 2 * sizeof(uintptr_t))) { + *flags |= CPU_DTRACE_BADADDR; + continue; + } + /* Get the size from the memref. */ size = memref[1];
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?202309061401.386E1X2p030800>